SoCalGas unveils branded, public-access compressed natural gas refueling station
Nissan expanding “No Charge to Charge” program; all-access EZ-Charge card

Promising Delphi 1st-gen Gasoline Direct Injection Compression Ignition engine meeting ultra fuel efficient program targets

Researchers at Delphi Powertrain, in collaboration with colleagues at Hyundai Motor, the University of Wisconsin-Madison, and Wisconsin Engine Research Consultants (WERC), have developed a first-generation multi-cylinder Gasoline Direct Injection Compression Ignition (GDCI) engine, based on several years of extensive simulations and single-cylinder engine tests. (Earlier post, earlier post.)

In a presentation at the SAE 2014 High Efficiency IC Engine Symposium and then in a paper given at SAE 2014 World Congress, Mark Sellnau, Engineering Manager, Delphi Advanced Powertrain, reported that Brake Specific Fuel Consumption for the 1.8L GDCI engine was significantly better than advanced production spark-ignited gasoline engines, and comparable to very efficient hybrid vehicle engines at their best efficiency conditions (214 g/kWh). Compared to new diesel engines, the Delphi team found that BSFC for GDCI at light loads was comparable or better, and at high loads was about 5% higher.

They found that at all loads, GDCI was remarkably clean, with the potential for no aftertreatment for NOx and particulate emissions. The team expects to be able to deliver further improvements in fuel consumption through planned development work.

Background. In 2010, The US Department of Energy (DOE) selected Delphi, along with partners Hyundai America Technical Center, Inc (HATCI); Wisconsin Engine Research Consultants (WERC); and the University of Wisconsin-Madison (UW) for a $7.48-million grant to develop and to demonstrate a new ultra fuel efficient vehicle (UFEV) vehicle concept. (Earlier post.) A key strategy for achieving the project goals was the further development of a new low-temperature combustion system: gasoline direct-injection compression-ignition (GDCI).

General project targets were to deliver diesel-like or better fuel economy using E10 gasoline (RON91), and to achieve low engine-out NOx and PM using low temperature combustion, the intention being to avoid the cost of aftertreatment for NOx and PM, in addition to meeting more stringent standards.

DOE UFEV program targets
Vehicle fuel efficiency on FTP +35% (PFI baseline)
ISNOx < 0.2 g/kWh over map
FSN (smoke number) < 0.1 (0.05) over map
CNL (combustion noise) load & speed dependent
Combustion stability < 5% COV IMEP (3%)

Throughout the world, many efforts are being made to improve the thermal efficiency of internal combustion engines. One relatively new approach is gasoline partially-premixed compression ignition (PPCI) that was introduced by Kalghatgi and first tested by Johansson. A high octane fuel was injected late on the compression stroke of a boosted diesel engine operating with high EGR. The injection process was complete prior to the start of combustion enabling partial mixing of the fuel and air prior to heat release. Very low fuel consumption, NOx, and PM emissions were measured. This early work established that gasoline-like fuels with high resistance to autoignition are preferred for PPCI.

… PPCI has demonstrated very good potential for very high fuel efficiency with low engine-out NOxand PM emissions using a range of gasoline-like fuels. However, towards a production solution, significant issues remain. Due to the lower exhaust enthalpy of lower temperature engines using PPCI, it is difficult to produce intake boost with acceptable boost system parasitis. A practical powertrain system with robust PPCI combustion is needed, including injection, valvetrain, boost, and exhaust subsystems. The engine must also meet vehicle packaging requirements under hood and satisfy cold start and transient response requirements.

… Delphi has been developing a multi-cylinder engine concept for PPCI combustion with the current US market gasoline (RON91).… A multiple-late-injection (MLI) strategy with GDI-like injection pressures was selected without use of a premixed charge. The absence of classic knock and pre-ignition limits in this process enabled a higher compression ratio of 15. The engine operates “full time” over the entire operating map with partially premixed compression ignition. No combustion mode switching, diffusion controlled combustion, or spark plugs were used. Delphi uses the term Gasoline Direct Injection Compression Ignition (GDCI) in reference to this combustion process.

—Sellnau et al. 2014-01-1300

Delphi says that GDCI combines the best of diesel and spark-ignited engine technology. Click to enlarge.

Current study. In this phase of the project, the engineers designed and fabricated a new multi-cylinder GDI engine, and used design tools to package the powertrain in a D-class vehicle. The conducted engine dynamometer tests over a range of operating conditions, and included preliminary calibration mapping.

They used this data for a competitive assessment of BSFC against published data for gasoline, diesel and hybrid engines.

They also simulated aggressive transients with a high rate of increased load as well as cold starts, and then tested all on the real engine.

GDCI engine. The GDCI engine features a shallow pent room combustion chamber, central-mounted injector, and 15:1 compression ratio. The quiescent, open-chamber design supports injection-controlled mixture stratification; swirl, tumble and squish were minimized to prevent destruction of the stratification created during injection. The piston and bowl—a symmetrical shape centered on the cylinder and injector axes—are matched to the injection system and spray characteristics.

The injection strategy is central to the success of the concept, and involves one, two or three injections during the compression stroke. At start of combustion (SOC), the stratified fuel-air mixture achieves stable combustion and controlled heat release.

The absence of classic combustion knock and preignition for GDCI means that a GDCI engine can be operated on RON91 gasoline at high compression ratio for high efficiency. For this reason, GDCI is a good candidate for aggressive downsizing, downspeeding, and boosting, which is a proven strategy for high vehicle-level fuel economy. These were two main consideration in base engine design.

—Sellnau et al.

Full-time GDCI combustion is achieved using exhaust rebreathing at low loads and cooled EGR at medium- to high-loads. Rebreathing is implemented using the exhaust valvetrain system, which provides a secondary exhaust lift event during the intake stroke. The team noted that this is an effective method to recuperate heat from hot exhaust gases in order to raise mixture temperatures—i.e., the heat promotes autoignition at low loads when boost pressure is zero or low.

Rebreathing also keeps the oxidation catalyst warm over a wide range of low loads without special maintenance heating.

A Delphi electric camshaft phaser actuates exhaust valvetrain and controls the secondary valve lift with very fast response.

The specially developed boost system (reported at the SAE events in 2013), features a turbocharger, supercharger, two intercoolers, and a low-pressure loop EGR system. The aftertreatment system consists only of an oxidation catalyst; tailpipe NOx and PM are to be handled by the GDCI combustion. The OxiCat is located at turbocharger outlet for heat conservation.

Schematic of boost system and OxiCat for aftertreatment. Click to enlarge.

Results. Engine dyno tests were run at idle, part-load, and full-load. For all operating conditions, the GDCI engine was operated with PPCI without mode switching or diffusion controlled combustion. Injection quantities and timings were used to control mixture stratification and combustion phasing. They found:

  • At idle and low load, rebreathing of hot exhaust gases provided stable combustion with NOx and PM emissions below targets of 0.2g/kWh and FSN 0.1, respectively. The coefficient of variation of IMEP was less than 3%. BSFC of 280 g/kWh was measured at 2000 rpm-2bar BMEP.

  • At medium-to-higher loads, rebreathing was not used and cooled EGR provided NOx, PM, and combustion noise below targets.

  • At full load operating conditions, near stoichiometric mixtures were used with up to 45% EGR. Maximum BMEP was about 20 bar at 3000 rpm.

  • Good BSFC at low speed and loads, and good minimum BSFC at higher load (~213 g/kWh).

  • Engine-out NOx was below 0.2 g/kWh at every point on the map—unique GDCI characteristic, they said. NOx decreases with increasing load, a helpful attribute for down-speeding.

  • Smoke (FSN) achieved program targets.

For the competitive assessment, they matched the GDCI’s data against:

  • BMW-PSA 1.6L Valvetronic 2013
  • Daimler 1.6L Strat. Camtronic 2012
  • Toyota 1.8L 3rd Gen Hybrid Atkinson-EGR
  • Honda 2.0L Hybrid Atkinson VTEC 2013
  • VW 2.0L Jetta Diesel T2B5 2009
  • Honda 2.2L i-DTEC Diesel 2012

They found:

  • For the best efficiency point on the map, the GDCI engine matched the hybrid vehicle engines; was about 5% higher than the two diesels; and was significantly beter than the spark-ignited engines.

  • For the low speed/load range, BSFC for GDCI is comparable to the hybrid engines; better than most of the diesels; and significantly beter than the SI data.

  • At the 2000 rpm, 2 bar BMEP world test point, BSFC for the GDCI was 280 g/kWh (clutch on), and 267 g/kWh (clutch off): lower than all the gasoline and diesel engines.

  • For the medium-to-high speed/load range, BSFC for GDCI was within 1 to 5% of diesel, and significantly beter than SI. The hybrid engines did not produce enough torque at these loads for comparison, and were labelled “off map.”

Despite the good results, the team emphasized that the results are not optimized; further improvements can come from further development of the combustion system, improving firing friction, and refined calibrations.


  • A, M., Foster, M., Hoyer, K., Moore, W. et al. (2014) “Development of a Gasoline Direct Injection Compression Ignition (GDCI) Engine,” SAE Int. J. Engines 7(2) doi: 10.4271/2014-01-1300



Interesting comparison to other engines, especially the part where both GDCI and hybrid engines exceeded diesel engine ;) This was probably due to the fact that the sweet spot for GDCI (and hybrid) is in lower power range than those two 2.0 diesels, maybe they should compare it to 1.4 diesel.


This could be the future of ICE. Gasoline may be a better fuel for CI engines if you look at the research of Kalghatgi, Johansson, and Rolf Reitz (U. of Wisconsin). Plus this engine is not that much different from existing ICE which use direct injection fuel systems, so there would not be a great cost to add a hybrid system (compared to a standard diesel engine).


I had a look at the SAE paper, not only the GCC article, and I will hint about a potential “show stopper” for this engine. The problem: this engine would have to be built much stronger than a diesel engine for similar torque and power. Explanation: At 17 bar BMEP, the maximum cylinder pressure is 160 bar. First, this is almost 10:1 ratio between the pressures. Second, in general, 17 bar in BMEP is not impressive at all for a turbo/supercharged engine. Recall that a state-of-the-art LD diesel engine today will achieve ~30 bar BMEP and 200 bar maximum cylinder pressure (pmax). This gives a factor less than 7:1. Note that lower is better, as long as it does not influence efficiency. Back in 1989, the first Audi TDI engine had a factor of 10:1, i.e. 135 bar and 13.5 bar respectively. It took more than a couple of years (~25 years) to reach today’s development level. For sure, we know more today about how to build an engine for high cylinder pressures but if this is the starting point, they will have a very, very long way to go. In the SAE paper, Delphi shows test results from 1500 to 2500 rpm, nothing higher. In simulations, they have 4000 rpm for rated power. Consequently, this is a normal speed range for a diesel engine and we should compare this engine with such an engine. If it cannot rev higher than a diesel, it should achieve (at least) similar BMEP levels. This cannot be achieved if cylinder pressure approaches 300 bar, as it would be with the factor 10:1. We simply do not know how to build such an engine yet. Scaling up the engine by almost a factor of two – or increasing the number of cylinders by a factor of two – to achieve similar torque and power as a conventional diesel engine would imply a substantial penalty in part load fuel consumption, not to mention the increase in physical size and weight. In summary, this concept has a long way to go before it could be production-ready.

Roger Pham

Good point, Peter.
The problem may be this:

"At full load operating conditions, near stoichiometric mixtures were used with up to 45% EGR. Maximum BMEP was about 20 bar at 3000 rpm."

Too much EGR! That is perhaps the reason why Daimler-Benz decided to go the SCR route instead of the high EGR, LTC route for NOx control. Perhaps use less EGR, homogeneous charge at high load and high speeds, and route the exhaust thru a TWC may solve the problem and produce more specific power and torque.

Roger Pham

On second thought, perhaps CR of 15 is too high for homogeneous mixture at low EGR and RON of 91. Perhaps a second fuel tank having higher octane fuel may be necessary for use in this high power mode to maximize power for maximal downsizing.

Roger Pham

On a third thought, scratch out the homogeneous charge because peak pressures would be too high with compression ignition, hence a second fuel tank would not be necessary. Thus at high power mode, back to PPCI, but with lower EGR rate, using stoichiometric mixture and a TWC for NOx control.


Don't forget the obvious, gasoline is lighter than diesel. So, even though BSFC numbers (fuel consumption by weight) are similar to diesel, the gasoline engine will consume a higher volume of fuel for a given output.


Thanks for the insider's critique, Peter XX.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)