Andreas Vressner

Ion Current Sensing for HCCI Combustion Feedback

Measurement of ion current signal from HCCI combustionwas performed. The aim of the work was to investigateif a measurable ion current signal exists and if it is possible to obtain useful information about the combustion process. Furthermore, influence of mixture quality in termsof air/fuel ratio and EGR on the ion current signal wasstudied. A conventional spark plug was used as ionizationsensor. A DC voltage (85 Volt) was applied acrossthe electrode gap. By measuring the current through thegap the state of the gas can be probed. A comparisonbetween measured pressure and ion current signal wasperformed, and dynamic models were estimated by usingsystem identification methods.The study shows that an ion current signal can be obtainedfrom HCCI combustion and that the signal levelis very sensitive to the fuel/air equivalence ratio. Themost important result from this study is that the ion current signal proved to be an excellent indicator of the actual combustion timing which is crucial piece of information for HCCI control.

Pressure Oscillations During Rapid HCCI Combustion

This work has focused on studying the in-cylinder pressure fluctuations caused by rapid HCCI combustion and determine what they consist of. Inhomogeneous autoignition sets up pressure waves traversing the combustion chamber. These pressure waves induce high gas velocities which causes increased heat transfer to the walls or in worst case engine damage. In order to study the pressure fluctuations a number of pressure transducers were mounted in the combustion chamber. The multi transducer arrangement was such that six transducers were placed circumferentially, one placed near the centre and one at a slight offset in the combustion chamber. The fitting of six transducers circumferentially was enabled by a spacer design and the two top mounted transducers were fitted in a modified cylinder head. During testing a disc shaped combustion chamber was used. The results of the tests conducted were that the in-cylinder pressure experienced during rapid HCCI-combustion is inhomogeneous. Pressure oscillations were experienced which showed good accordance to vibration mode shapes and frequencies suggested by acoustic vibration theory. The pressure waves manifested largest intensities for the first vibration mode, a mode suggesting radial propagation of the pressure waves in the combustion chamber. Experiments showed that the direction of the pressure wave was random which hinted absence of hot spot ignition. Hot spots mean that some part of the combustion chamber is physically hotter than the rest which ignites adjacent mixture. There is no evidence available yet to demonstrate engine damage during rapid HCCI combustion. An explanation to this could be that the violent combustion reactions related to SI knock damage is not present in HCCI combustion due to the diluted mixtures. The local heat released will therefore be lower. Finally, engine tests using two other combustion chamber geometries were conducted. The results showed that altering the geometry of the combustion chamber affects the resulting frequency spectrum. The two geometries were hill- and a bowl shaped respectively. Analytical calculations on the bowl shape vibration frequencies indicate reasonably good accordance to experimental results.

Optical Diagnostics of Laser-Induced and Spark Plug-Assisted Hcci Combustion

HCCI (Homogeneous Charge Compression Ignition), laser-assisted HCCI and spark plug-assisted HCCI combustion was studied experimentally in a modified single cylinder truck-size Scania D12 engine equipped with a quartz liner and quartz piston crown for optical access. The aim of this study was to find out how and to what extent the spark, generated to influence or even trigger the onset of ignition, influences the auto-ignition process or whether primarily normal compression-induced ignition remains prevailing. The beam of a Q-switched Nd:YAG laser (5 ns pulse duration, 25 mJ pulse energy) was focused into the centre of the cylinder to generate a plasma. For comparison, a conventional spark plug located centrally in the cylinder head was alternatively used to obtain sparks at a comparable location. No clear difference in the heat releases during combustion between the three different cases of ignition start could be seen for the fuel of 80/20 iso-octane/n-heptane used. However, with optical diagnostic methods, namely PLIF (Planar Laser-Induced Fluorescence), Schlieren photography and chemiluminescence imaging, differences in the combustion process could be evaluated.

Multiple Point Ion Current Diagnostics in an HCCI Engine

Interest in ion current sensing for HCCI combustion arises when a feedback signal from some sort of combustion sensor is needed in order to determine the state of the combustion process. A previous study has revealed that ion current sensors in the form of spark plugs can be used instead of expensive piezoelectric transducers for HCCI combustion sensing. Sufficiently high ion current levels were achieved when using relatively rich mixtures diluted with EGR. The study also shows that it is not the actual dilution per se but the actual air/fuel equivalence ratio which is important for the signal level. Conclusions were made that it is possible to obtain information on combustion timing and oscillating wave phenomena from the measurements. However, the study showed that the ion current is local compared to the pressure which is global in the combustion chamber. This observation triggered the present study where the aim is to investigate the ion current at different locations in the combustion chamber. The ion current was measured simultaneously at seven locations in the combustion chamber. In order to achieve this, 6 spark plugs were fitted circumferentially in a spacer placed between the cylinder block and the head. The seventh spark plug was placed in the cylinder head. Individual DC sources of 85 volts were applied across the spark plug gaps. The present study indicates that the combustion timing seems to be dependent on the wall temperature at the different spark plug locations. The largest difference in timing between different locations in the combustion chamber was 2 CAD. The ion current amplitude varies with different spark plug locations up to 1.5 μA. The signal strength increases with decreasing air/fuel ratio and is also affected by dilution.

Effect of Turbulence on HCCI Combustion

This paper presents large eddy simulation (LES) and experimental studies of the combustion process of ethanol/air mixture in an experimental optical HCCI engine. The fuel is injected to the intake port manifolds to generate uniform fuel/air mixture in the cylinder. Two different piston shapes, one with a flat disc and one with a square bowl, were employed to generate different in-cylinder turbulence and temperature field prior to autoignition. The aim of this study was to scrutinize the effect of in-cylinder turbulence on the temperature field and on the combustion process. The fuel tracer, acetone, is measured using laser-induced fluorescence (LIF) to characterize the reaction fronts, and chemiluminescence images were recorded using a high-speed camera, with a 0.25 crank angle degree resolution, to further illustrate the combustion process. Pressure in the cylinder is recorded in the experiments. Spatial and temporal resolved LES was used to gain information on the turbulence mixing, heat transfer and combustion process. It was shown that gas temperature in the piston bowl is generally higher than that in the squish, leading to an earlier ignition in the bowl. Compared to the disc engine, the square bowl engine has a higher temperature inhomogeneity owing to the turbulence wall heat transfer. The experimentally observed higher combustion duration and slower pressure rise rate in the square bowl engine as compared to the disc engine can be explained by the higher temperature inhomogeneity in the square bowl engine.

Cycle Resolved Wall Temperature Measurements Using Laser-Induced Phosphorescence in an HCCI Engine

Cycle-resolved wall temperature measurements have been performed in a one-cylinder, port-injected optical Scania D12 truck engine run in HCCI mode. Point measurements at various locations were made using Laser-Induced Phosphorescence (LIP). Single point measurements with thermographic phosphors utilize the temperature dependency of the phosphorescence decay time. The phosphorescence peak at 538 nm from the thermographic phosphor La2O2S:Eu was used to determine temperature. A frequency tripled 10 Hz pulsed Nd:YAG laser delivering ultraviolet (UV) radiation at 355 nm was used for excitation of the phosphor. Detection in the spectral region 535 - 545 nm was performed every cycle with a photo multiplier tube connected to a 3 GHz oscilloscope. Measurements were made at four points on the cylinder head surface and two points on the outlet and inlet valves respectively. For each location measurements were made at different loads and at different crank angle degrees (CAD). The aim of the presented work was to study the feasibility of using LIP for single-shot, cycle-resolved wall temperature measurements. (Less)

Study on Combustion Chamber Geometry Effects in an HCCI Engine using High-Speed Cycle-Resolved Chemiluminescence Imaging

The aim of this study is to see how geometry generated turbulence affects the Rate of Heat Release (ROHR) in an HCCI engine. HCCI combustion is limited in load due to high peak pressures and too fast combustion. If the speed of combustion can be decreased the load range can be extended. Therefore two different combustion chamber geometries were investigated, one with a disc shape and one with a square bowl in piston. The later one provokes squish-generated gas flow into the bowl causing turbulence. The disc shaped combustion chamber was used as a reference case. Combustion duration and ROHR were studied using heat release analysis. A Scania D12 Diesel engine, converted to port injected HCCI with ethanol was used for the experiments. An engine speed of 1200 rpm was applied throughout the tests. The effect of air/fuel ratio and combustion phasing was also studied. The behavior of the heat release was correlated with high speed chemiluminescence imaging for both combustion chamber geometries. Optical access was enabled from beneath by a quartz piston and a 45 degree mirror. It was found that the square bowl in piston generates higher turbulence levels resulting in half the ROHR and twice as long combustion duration as the disc shaped combustion chamber. By using a resolution of 3 images per CAD, the fast gas movements during the entire HCCI combustion process could be studied inside the bowl.

Large Eddy Simulation and Experiments of the Auto-Ignition Process of Lean Ethanol/Air Mixture in HCCI Engines

Recent experiments and numerical studies have showed that piston geometry has a significant effect on the homogeneous charge compression ignition (HCCI) process. There are two effects generated by the combustor geometry: the geometry affects the flow/turbulence in the cylinder; the geometry also affects the temperature stratification. The temperature stratification is more directly responsible for the observed alteration of the auto-ignition process. To clarify this issue further we present in this paper a study of two engines with the same geometry but difference ways of cooling. Measurement of the two engines~a metal engine and quartz piston engine, both with the same piston bowl geometry~is carried out. Large eddy simulation (LES) is used to simulate the flow, the temperature field and the auto-ignition process in the two engines. The fuel is ethanol with a relative air/fuel ratio of 3.3. It is found that lower temperature stratification is established in the metal engine under similar conditions as the optical quartz engine due to the more effective cooling of the piston in the metal engine configuration. The combustion phasing in the two engines is the same by controlling the intake temperature. Both measurements and LES show a more rapid auto-ignition in the metal engine than in the optical engine with the same piston geometry. This confirms the conclusion that large temperature stratification can decrease the pressure-rise-rate and thereby increase the load of HCCI engines. The dependence of temperature stratification on the wall temperature and intake temperature is systematically studied using LES. (Less)

Fuel Effects on Ion Current in an HCCI Engine

An interest in measuring ion current in Homogeneous Charge Compression Ignition (HCCI) engines arises when one wants to use a cheaper probe for feedback of the combustion timing than expensive piezo electric pressure transducers. However the location of the ion current probe, in this case a spark plug, is of importance for both signal strength and the crank angle position where the signal is obtained. Different fuels will probably affect the ion current in both signal strength and timing and this is the main interest of this investigation. The measurements were performed on a Scania D12 engine in single cylinder operation and ion current was measured at 7 locations simultaneously. By arranging this setup there was a possibility to investigate if the ion current signals from the different spark plug locations would correlate with the fact that, for this particular engine, the combustion starts at the walls and propagates towards the center of the combustion chamber. The fuels investigated were isooctane, n-heptane, PRF80, gasoline, diesel, ethanol and methanol. A special interest was how the ion current timing was affected by low temperature reactions, which were present with the n-heptane and diesel fuels as well as mixtures of isooctane and n-heptane, i.e., PRF80. The most interesting results were that ion current is both affected by the ion current probe location in the combustion chamber and the fuel used. Fuels with higher octane numbers seem to provoke ion current more easily, thus with LTR fuels as n-heptane and diesel ion current was only achieved at richer mixtures. The cycle-to-cycle variations of ion current increased with leaner mixtures. Ion current was also affected by combustion phasing and engine speed. (Less)

High-Speed LIF Imaging for Cycle-Resolved Formaldehyde Visualization in HCCI Combustion

High-speed laser diagnostics was utilized for single-cycle resolved studies of the formaldehyde distribution in the combustion chamber of an HCCI engine. A multi-YAG laser system consisting of four individual Q-switched, flash lamp-pumped Nd:YAG lasers has previously been developed in order to obtain laser pulses at 355 nm suitable for performing LIF measurements of the formaldehyde molecule. Bursts of up to eight pulses with very short time separation can be produced, allowing capturing of LIF image series with high temporal resolution. The system was used together with a high-speed framing camera employing eight intensified CCD modules, with a frame-rate matching the laser pulse repetition rate. The diagnostic system was used to study the combustion in a truck-size HCCI engine, running at 1200 rpm using n-heptane as fuel. By using laser pulses with time separations as short as 70 μs, cycle-resolved image sequences of the formaldehyde distribution were obtained. Thus, with this technique it is possible to follow the formaldehyde formation and consumption processes within a single cycle. The combustion evolution was studied in terms of the rate and spatial structure of formaldehyde formation and consumption for different engine operating conditions, e.g. different stoichiometries. Also, the impact on the rate of heat-release was investigated.