Honda is striving to develop a form of lean burn with homogeneous premixture that would be able to balance low NO_x emissions with stable, controllable combustion. The Honda HLSI approach seeks to take the best of conventional lean-burn (controllability and low combustion noise) and homogeneous charge compression ignition (HCCI) (ultra-Low NO_x and high thermal efficiency) while avoiding the downsides of each: NO_x emissions for the former and operating range, controllability and combustion noise for the latter.

Honda envisions HLSI as delivering high thermal efficiency with less than 50 ppm engine-out NOx and low combustion noise. Click to enlarge.

A great deal of research has been conducted on lean burn as an effective means of increasing the efficiency of gasoline engines. In a lean burn engine, three-way catalysts show virtually no NO_x purification effect. In order to achieve a sufficient reduction in engine-out NO_x emissions, it is therefore necessary to increase the combustion limit air-fuel ratio (termed the “lean limit” below).

In conventional stratified-charge lean combustion engines, because a rich air-fuel mixture is generated in the vicinity of the spark plugs, NO_x is emitted in the first half of the combustion. In stoichiometric combustion, close to 100% of the NO_x is purified by the three-way catalyst, but in lean burn the three-way catalyst is almost completely incapable of NO_x reduction. Thus, for the emission control value to be satisfied, the NO_x must be trapped and reduced using a lean NO_x catalyst (LNC). When the NO_x in the LNC reaches the permissible amount, lean burn is stopped, and rich combustion is performed to reduce the NO_x. As a result, the cycle for this rich combustion reduces the fuel efficiency. As emission control values have become stricter year after year, NO_x reduction cycles using rich combustion have also increased, and there are now almost no automobiles with lean burn engines being sold.

Considerable research has been conducted on HCCI combustion, in order to resolve this problem, but due to factors including the narrowness of the potential operating range and the challenge represented by controllability in response to fluctuations in the environment this method has not yet entered mass production.

To improve fuel-efficiency and cost, this study aims to contribute to the construction of lean burn technology that does not require an LNC. The research discussed in this paper therefore focused on a form of lean burn involving the stable spark ignition of a homogeneous premixture that would be able to balance low NO_x emissions with combustion controllability. To that end, the primary goal of the study is to identify the mechanism by which the limit of homogeneous lean burn, which can contribute to reduced NO_x emissions, is reached. Building on this knowledge, the research aims to achieve a high-efficiency, low-NO_x homogeneous lean burn engine.

—Hanabusa et al.

Homogeneous lean premixed gas is known to have poor flame propagation characteristics. To determine the dominant cause of this, the Honda study investigated the combustion properties of a single-cylinder engine while changing the compression ratio and intake temperature.

In its initial set of experiments, Honda modified a single-cylinder experimental engine so that fuel was injected from a from an upstream PI (port-injector) approximately 1 m above the engine intake valve such that it mixed homogeneously with air. The fuel-air mixture was heated by the heater and supplied to the engine. Honda experimented with different compression ratio and intake air temperature combinations to investigate what factors determined the lean limit.

The researchers found that abnormal cycles—the primary cause of combustion fluctuation—had low TDC temperatures compared to those of other cycles. Cycles with an unburned gas temperature in TDC below 950 K (677 ˚C) had particularly poor heat production during the expansion stroke.

They also found that laminar burning velocity increased rapidly when the temperature of the unburned mixture exceeded approximately 1,000 K (727 ˚C). Their investigation showed that when the temperature of the unburned mixture reached 1,000 K or more, the decomposition of H₂O₂ produced OH radicals, which promoted combustion.

To evaluate the applicability of the combustion concept to mass-production engines, Honda then used a single-cylinder engine derived from a mass-produced 4- cylinder engine for which the temperature at TDC would exceed 1,000 K . Although Honda maintained the engine’s port injection system, the researchers switched to an atomization nozzle to make the mixture as homogeneous as possible.

To increase the heat production of the lean mixture before TDC when the laminar burning velocity is slow, this engine featured a different compression ratio, intake system, piston shape, and ignition system.

“Heat loss (other)” refers to heat loss other than that caused by cooling water. Click to enlarge.

Modifications made to the engine resulted in an increase in the lean limit to A/F 31. Further, the highest efficiency combustion was achieved at A/F 30; they also achieved less than 50ppm NO_x emission. The engine was operated at an engine speed of 1500 rpm and an NMEP of 500 kPa.

HLSI combustion increased thermal efficiency by 6.2 points over stoichiometric combustion, from 33.7% to 39.9% at A/F 30. Honda attributed reductions in cooling loss (resulting from low in-cylinder gas temperature of HLSI) and pumping loss (due to significant air dilution) as the main causes of the increased efficiency. Exhaust loss and unburned fuel loss became the main obstacles to efficiency.

Based on work to date, Honda has broadly concluded that:

• Providing an SI engine with a homogeneous premixed gas and adjusting the compression ratio and intake temperature so that the in-cylinder temperature increases results in a higher lean limit and also results in extremely low NO_x emissions.

• The compression ratio and intake air temperature combination have a significant impact on the lean limit, and combinations that lead to higher in-cylinder temperatures result in higher lean limits. Also, the engine-out NO_x is lower for combinations that lead to higher in-cylinder temperatures.

• When TDC in-cylinder temperature reaches approximately 1,000 K (727 ˚C), H₂O₂ decomposition reactions occur, and as a result, the laminar burning velocity increases and combustion fluctuation is stabilized.

• The analysis of the fractions of fuel energy in homogeneous lean mixture combustion has clarified the mechanism by which thermal efficiency increases.

• The combustion concept holds good in a mass-production engine and confirmed that both high thermal efficiency and low NO_x emissions can be achieved in flame propagation.

While the engine-out NO_x levels are low, they are still insufficient to reach the coming standards in Japan and the US, Kondo noted. Thus, some aftertreatment would be required.

Resources

• Hanabusa, H., Kondo, T., Hashimoto, K., Sono, H. et al. (2013) “Study on Homogeneous Lean Charge Spark Ignition Combustion,” SAE Technical Paper 2013-01-2562 doi: 10.4271/2013-01-2562