Ricardo launches pre-production prototype of high-speed flywheel energy storage for off-road equipment
Sandia Labs project team building fuel cell cold ironing system for deployment at Port of Honolulu in 2015

SwRI develops D-EGR demonstrator highlighting fuel-efficient, cost-effective engine; 10% and above fuel economy improvements

Southwest Research Institute (SwRI) has built a demonstrator vehicle, based on a 2012 Buick Regal GS, incorporating the latest D-EGR (dedicated EGR) technology (earlier post), an outgrowth of SwRI’s HEDGE (High Efficiency Dilute Gasoline Engine) consortium projects (earlier post). The vehicle is currently delivering about a ~13% improvement on the FTP and 10% on the HwFET cycles, according to Dr. Terry Alger, SwRI Assistant Director, Engine & Vehicle R&D, Engine, Emissions & Vehicle Research.

Staff at the non-profit research organization presented it with little fanfare to the SwRI board of directors and advisory trustees along with 7 other research projects at the 66th annual meeting on Monday. The demonstrator will be present at the upcoming SAE 2014 High Efficiency IC Engine Symposium in Detroit in April; SwRI engineers are presenting a paper (2014-01-1190) at the SAE 2014 World Congress on the development of the demonstrator’s engine.

D-EGR is an engine and control concept that combines fuel reforming to produce hydrogen with high levels of exhaust gas recirculation (EGR) to achieve very high levels of thermal efficiency. In 2013, SwRI reported a 2.0-liter engine running as a full D-EGR engine from idle to full load, delivering more than 42% brake thermal efficiency with ultra-low exhaust emissions.

To accomplish reformation of the gasoline in a cost-effective, energy efficient manner, a dedicated cylinder is used for both the production of EGR and reformate. By operating the engine in this manner, many of the sources of losses from traditional reforming technology are eliminated and the engine can take full advantage of the benefits of reformate.

As will be presented at SAE World Congress, SwRI engineers modified the 2.0-liter engine in the demonstrator to add:

  • the dedicated EGR loop;
  • an additional injector for delivering extra fuel for reformation;
  • a modified boost system that included a supercharger;
  • high energy dual coil offset (DCO) ignition; and
  • other actuators used to enable the control of D-EGR combustion.

The compression ratio of the engine was increased to 11.7:1 to take advantage of the improved knock resistance from reformate and EGR.

The conversion to D-EGR improved engine efficiency by at least 10% across the performance map, with some operating conditions seeing substantially higher improvements, as noted above.

The SwRI team will report that the brake specific fuel consumption (BSFC) at 2000 rpm 2 bar BMEP improved from 385 g/kW-h in the series production state to 330 g/kW-h; the lowest BSFC for the engine was 212 g/kWh compared to 236 g/kW-h for the series engine.

HEDGE III_wTM-220pixels

The addition of 2-stage boosting also allowed the engine to meet its torque targets of at least 17 bar BMEP from 1500-5500 rpm while maintaining good transient response and low engine-out emissions.

Last February, PSA Peugeot Citroën announced that it will commercialize high-efficiency gasoline engines featuring dedicated exhaust gas recirculation (D-EGR). At that time, PSA said it expected the new D-EGR engines, due to be available in PSA vehicles by 2018, would consume 10% less fuel than their predecessors across all uses. PSA has been a member of the HEDGE consortium (now at HEDGE-III) since 2004.


  • Christopher Chadwell, Terrence Alger, Jacob Zuehl, Raphael Gukelberger (2014) “A Demonstration of Dedicated EGR on a 2.0 L GDI Engine,” (SAE 2014-01-1190)



10% savings isn't a huge amount, but at reduced cost?  This ought to be going into everything as fast as humanly possible.


Just reading the abstract for the paper (I'm too cheap to spend the $17 or so to see the whole story). It's intriguing. I am most skeptical with respect to transient response. It's difficult for me to see how the reformate composition can be optimally maintained during large transients. The language of the abstract ("while maintaining good transient response") seemed a little tepid -- sort of like the real estate ads whose most positive trait is "cozy".

Still, the numbers are impressive. The paper claims SFC as low as 212g/kWh. If you assume the 2012 Regal GS being used in the test needs 15kW for a 100km/h cruise, that would equate to a 4.24l/100km or roughly 55mpg consumption on gasoline. Very interesting.

Roger Pham

15 kW is only 1/7th of the peak output of a typical 2.0 liter NA engine of over 100 kW. The Buick Regal GS is turbocharged and is capable of 270 hp (200 kW). As such, the BSFC won't be close to the lowest possible number of .212, but rather much closer to .330kg/kWh. So, a revised number in term of this adjustment would give a number closer to 35 mpg. Another way to look at this is that the measured improvement in efficiency is around 10% vs a baseline engine. A 2012 Buick Regal GS has an EPA hwy mileage of 27 mpg. Increase it by 10% would give it about 30 mpg.

Perhaps cost reduction in comparison to Diesel engine with urea-SCR and DPF and high-pressure common rail injectors...but some additonal cost will incur in comparison to a base Buick Regal GS turbocharged engine, as outline in the article.

Still, a pretty ingenious way to eke out further efficiency in the gasoline ICE.


Intuitively, adding combustion exhaust to fresh air intake doesn't necessarily sound like an engine cost or efficiency improvement.

Roger Pham

EGR (Exhaust Gas Recirculation, or recycling some of the exhaust gas back into the intake) is already done in current ICE in order to meet NOx emission regulation. This is done at less than maximum power in order to lower the combustion temperature by diluting the intake charge without having excess oxygen that will render the three-way catalytic converter ineffective. EGR is not used at maximum power due to the slowness of combustion, however , is possible here due to the presence of H2.

The ingenuity here is to dedicate only one cylinder for that purpose, and to run that cylinder rich in order for the excess fuel to be reformed into H2 and CO. This has four advantages:
1. reformation process is endothermic, meaning that some exhaust waste heat can be recycled,
2. the presence of H2 will increase the speed and completeness of combustion in diluted charge that would otherwise burn slower and will limit top power, and dilute charge will allow increase in compression ratio that will allow higher thermal efficiency,
3. exhaust emission will be significantly improved with diluted charge and rapid combustion, when the combustion is at lower temperature, thus shifting the equilibrium away from NOx formation, and the O2 is rapidly consumed leaving much less chances of forming NOx, and CO and HC will also be lower in the presence of H2,
4. EGR can be used even for maximum power production, thus allowing higher compression ratio and better post-combustion emission.

In fact, the improved emission in order to satisfy future emission regulations by itself would justify this technology, in order to make ICEV's possible way into the future, when the EPA and CARB are trying to squeeze them out by keep tightening the snooze, hoping to strangle ICEV's. The efficiency improvement is a nice bonus!


RP, if all is as stated it sounds good.


I think you missed this:
5.  To run with both dilute mixtures and stoichiometric combustion (required for a TWC), the intake air must be mixed with something that is essentially oxygen-free.  EGR fills the bill, EGR from a cylinder used for reforming fuel to a fast-burning mixture is even better.

I keep coming back to Transonic Combustion.  The exhaust from the reforming cylinder would be perfect for fuel diluent, and injecting supercritical fuel would recycle even more exhaust heat.  Why haven't these things been put together yet?

william g irwin

It sounds like Peugeot Citroen is considering commercialization as we chat. Who's next - Ford maybe? Then we'll know just how practical this is. Great technology transfer coming soon!!

The comments to this entry are closed.