Homogeneous charge compression ignition (HCCI) engines lack direct in-cylinder mechanisms, such as spark plugs or direct fuel injection, for controlling the combustion timing. Many indirect methods have been used to control the combustion timing in an HCCI engine. With any indirect method, it is important to have a measure of the combustion timing so the control inputs can be adjusted for the next cycle. In this paper, it is shown that microphones and knock sensors can be used to detect combustion in HCCI engines. The output from various microphones and a knock sensor on an HCCI engine are measured at light and high loads. The combustion timing data obtained from the sensors are compared to the combustion timing data obtained from a piezoelectric cylinder pressure transducer. One of these sensors is selected and used for closed-loop control of the combustion timing in a single cylinder HCCI engine.

This demonstration system is intended to meet the California Energy Commission’s primary goal of improving California’s electric energy cost/value by providing a low-cost high-efficiency distributed power generation engine that runs on landfill gas. The project team led by Makel Engineering, Inc. includes UC Berkeley, CSU Chico and the Butte County Public Works Department.

The team has developed a reliable, multi-cylinder Homogeneous Charge Compression Ignition (HCCI) engine by converting a Caterpillar 3116, 6.6 liter diesel engine to operate in HCCI mode. This engine utilizes a simple and robust thermal control system. Typically, HCCI engines are based on standard diesel engine designs with reduced complexity and cost based on the well known principles of engine dynamics. Coupled to an induction generator, this HCCI genset allows for simplified power grid connection.

Testing with this HCCI genset allowed for the development of a control system to maintain optimal the inlet temperature and equivalence ratio. A brake thermal efficiency of 35.0% was achieved while producing less than 10.0 ppm of NOx and 30 kW of electrical power. Less than 5.0 ppm of NOx was recorded with a slightly lower brake thermal efficiency. Tests were conducted with both natural gas and simulated landfill gas as a fuel source. This demonstration system has shown that landfill gas fueled Homogeneous Charge Compression Ignition engine technology is a viable technology for distributed power generation.

Interest in the use of alternative fuels and engines is increasing as the price of petroleum climbs. The inherently higher efficiency of Diesel engines has led to increased adoption of Diesels in Europe, capturing approximately 40% of the new passenger car market. Unfortunately, lower CO2 emissions are countered with higher nitrogen oxides (NOx) and particulate matter (PM) emissions, and higher noise. Noise and PM have traditionally been the obstacles toward consumer acceptance of Diesel passenger cars in North America, while NOx (a key component in photochemical smog) has been more of an engineering challenge. Diesels have non-premixed combustion with excess oxygen; reducing NOx to N2 in an oxygen rich environment is difficult. Adding oxygenated compounds to the fuel helps reduce PM emissions. However, relying on fuel alone to reduce PM is unrealistic due to economic constraints and difficult due to the emerging PM standards. Keeping peak combustion temperature below 1700 K inhibits NOx formation. Altering the combustion regime to burn at temperatures below the NOx threshold and accept a wide variety of fuels seems like a promising alternative for future engines. Homogeneous Charge Compression Ignition (HCCI) is a possible solution. Fuel and air are well mixed prior to intake into a cylinder (homogeneous charge) and ignition occurs by compression of the fuel-air mixture by the piston. HCCI is rapid and relatively cool, producing little NOx and PM. Unfortunately, it is hard to control since HCCI is initiated by temperature and pressure instead of a spark or direct fuel injection. We investigate biofuel HCCI combustion, and use intrinsically labeled biofuels as tracers of HCCI combustion. Data from tracer experiments are used to improve our combustion modeling.

The influence of the small amounts (1-3%) of the additive di-tertiary butyl peroxide (DTBP) on the combustion event of Homogeneous Charge Compression Ignition (HCCI) engines was investigated using engine experiments, numerical modeling, and carbon-14 isotope tracing. DTBP was added to neat ethanol and diethyl ether (DEE) in ethanol fuel blends for a range of combustion timings and engine loads. The addition of DTBP to the fuel advanced combustion timing in each instance, with the DEE-in-ethanol mixture advancing more than the ethanol alone. A numerical model reproduced the experimental results. Carbon-14 isotope tracing showed that more ethanol burns to completion in DEE-in-ethanol blends with a DTBP additive when compared to results for DEE-in-ethanol without the additive. However, the addition of DTBP did not elongate the heat release in either case. The additive advances combustion timing for both pure ethanol and for DEE-in-ethanol mixtures, but the additive results in more of an advance in timing for the DEE-in-ethanol mixture. This suggests that although there are both thermal and kinetic influences from the addition of DTBP, the thermal effects are more important.

Homogeneous Charge Compression Ignition (HCCI) engines are amenable to a large variety of fuels as long as the fuel can be fully vaporized, sufficiently mixed with air, and receive sufficient heat during the compression stroke to reach the autoignition conditions. This study investigates an HCCI engine fueled with ethanol-in-water mixtures, which we call “wet ethanol”. The motivation for using wet ethanol fuel is that significant energy is required for distillation and dehydration of fermented ethanol (from biosources, not from petroleum), thus direct use of wet ethanol could improve energy balance. Recent modeling studies have predicted that a HCCI engine can operate using fuel containing as little as 35% ethanol-in-water, with surprisingly good performance and emissions. With the previous modeling study suggesting feasibility of wet ethanol use in HCCI engines, this paper focuses on experimental operation wet ethanol in a 4-cylinder 1.9 liter engine running in HCCI mode. This study investigates the effect of the ethanol-water fraction on the engine's operating limits, intake temperatures, heat release rates, and exhaust emissions for the engine operating with 100%, 90%, 80%, 60%, and 40% ethanol-in-water mixtures.

Despite the rapid combustion typically experienced in Homogeneous Charge Compression Ignition (HCCI), components in fuel mixtures do not ignite in unison or burn equally. In our experiments and modeling of blends of diethyl ether (DEE) and ethanol (EtOH), the DEE led combustion and proceeded further toward completion, as indicated by 14C isotope tracing. A numerical model of HCCI combustion of DEE and EtOH mixtures supports the isotopic findings. Although both approaches lacked information on incompletely combusted intermediates plentiful in HCCI emissions, the numerical model and 14C tracing data agreed within the limitations of the single zone model. Despite the fact that DEE is more reactive than EtOH in HCCI engines, they are sufficiently similar that we did not observe a large elongation of energy release or significant reduction in inlet temperature required for light-off, both desired effects for the combustion event. This finding suggests that, in general, HCCI combustion of fuel blends may have preferential combustion of some of the blend components.

In lean premixed combustion systems, inadequate mixing of the fuel and air, prior to combustion can cause unnecessarily large pollutant emissions. Measuring the extent of mixing of fuel into air is often difficult, since combustion in lean premixed gas turbines takes place at high pressures, often making optical access to the combustion area limited. In addition, the pressure broadening of the molecular absorption lines renders the spectrally narrow line associated with a laser light source less useful. This paper studies some of the problems in determining the extent of mixing of the fuel into air in these lean premixed combustion systems. The focus of this paper is the use of an infrared light emitting diode (IR-LED) to quantitatively measure fuel concentration in a lean premixed gas turbine. The IR-LED emits radiation over a wide wavelength range compared to a laser, meaning that the development of an absorption coefficient to relate the fuel concentration to the absorption of the IR-LED radiation is not as direct as developing the absorption coefficient for the absorption of laser light. Controlled experiments were performed where the pressure, path length and fuel concentration were varied and the effects of these three parameters on the absorption of radiation from the IR-LED were studied. A broad band absorption coefficient was developed relating the absorption of light from the IR-LED to the fuel concentration. This broad band absorption coefficient was found to be in good agreement with calculated coefficient values. Experiments were performed on a lean premixed gas turbine combustor modified for line-of-sight optical access. The concentration profile of this high pressure combustor was found by tomographic reconstruction from line-of-sight absorption measurements using the IR-LED. We demonstrated that the IR-LED can be used for quantitative measurements of the fuel concentration for high pressure systems.