Unraveling advanced compression ignition combustion using optical diagnostics
- Publisher: Technische Universiteit Eindhoven
Despite the expected upsurge of hybrid and electric cars in the coming decades, internal combustion will remain the main power supply for (long-distance) transport. Buses, trucks, ships and airplanes will still rely on combustion engines. Nevertheless, emission legislation is becoming more stringent and the oil price continues to rise. Consequently, there still exists a serious interest in new developments that may improve combustion efficiency and fuel flexibility, and reduce emissions; both fundamental and applied research in this field is thriving. Recently, a lot of research has focused on advanced compression ignition combustion strategies, like premixed charge compression ignition (PCCI). These strategies aim at combining the advantages of gasoline and diesel combustion within the same engine. As a result, engines will be more efficient and have extremely low exhaust emissions enabling to meet emission legislations without after treatment. To gain understanding of the physical processes occurring in the ‘black box’ of an engine, engines can be made optically accessible, thus allowing a look inside the engine during the combustion process. In these so-called "optical engines" various optical diagnostics can be used to visualize in-cylinder processes, and to supply data on e.g. flow fields and temperature. As a next step, empirically measured data can be used to optimize numerical calculations, which, in turn, hopefully will improve to such extent that they can be used to replace testing of real manufactured prototypes. One of the main topics in the area of internal combustion research is to identify and explain phenomena around the onset of ignition, including the injection of fuel itself. In the short time period between the start of fuel injection and ignition (typically a few milliseconds), all events of interest take place. Large cycle-to-cycle variations in real engines require fast diagnostic techniques to resolve these short relevant time spans in single combustion cycles. Therefore, fast optical diagnostic techniques are of major importance. In this thesis, novel optical diagnostics and improved combustion strategies for compression ignition internal combustion engines are presented. Each chapter focuses on a specific optical measurement technique, which can be used to characterize the different aspects of compression ignition combustion in detail. Velocities. To investigate mixing of fuel and air, the in-cylinder gas-flow velocities were measured using particle image velocimetry (PIV) in chapter 3 and 4. From these velocities, the turbulence intensity was derived, which gives an indication of the degree of mixing of the charge. These measurements have been performed in a 2D plane parallel to and a few mm below the cylinder head. The 2D measurements indicate that the turbulence is homogeneous and that the swirl center shifts towards the center of the combustion chamber during the compression stroke due to squish motion. Increasing the recording rate by using a high-speed laser in combination with a high-speed camera enables to investigate the flow evolution in a single cycle, and to record more data before window fouling occurs. To achieve enough contrast between the particles and the environment, the relatively low laser power of the high-speed laser system needs to be compensated by using hollow microspheres as tracers which have a larger size compared to the more generally used oil droplets. The spatial resolution of the presented in-cylinder velocity fields in chapter 4 is shown to be lower for the high-speed measurements than for the low-speed measurements presented in chapter 3. High-speed PIV is therefore not a substitute for high-resolution low-speed PIV, but an additional measurement method to track fast changes. The high-speed PIV results during and after injection results show a sudden change of air motion at the start of injection as a result of air entrainment at the core of the spray. Furthermore, as expected, spray injection causes a considerable increase in the cycle-to-cycle fluctuations of the flow pattern, the more so for longer injection durations. In the case of multiple injections in a single engine cycle the air-entrainment during the first injection is consistent, and fluctuations between consecutive cycles are small. When fuel vapor from previous injections re-enters the investigated plane via impingement on the cylinder wall or the top of the piston, the flow structure changes drastically and loses coherence, compensating the inward motion due to air entrainment by subsequent injections. The spray induced flow can evolve into various structures, which might influence the actual mixing of fuel and air, causing differences in local fuel/air ratios between different injection strategies and therefore change the ignition delay in PCCI combustion. The high-speed PIV data has been subjected to proper orthogonal decomposition analysis (POD). The results confirm the observed changes in flow structures during and after injection. POD might be a good tool for comparison of experimental PIV data with future CFD results. However, in highly unsteady flow, as observed during injection, a good representation of the instantaneous velocity field using only a few POD modes is not possible and therefore comparison between PIV and CFD data needs careful selection of the amount of modes. Phase-invariant POD was found not to be an appropriate method to represent highly fluctuating flows during and after fuel injections. Because of the lack of resemblance between flow fields of consecutive crank angles, the use of separate basis functions for each CAD is more appropriate; alternatively one could use separate sets of base functions before and during injection. Temperature gradients Temperature gradients occurring before, during and after fuel injection were measured using a 2-color laser induced fluorescence (LIF) technique with toluene as a fuel tracer (chapter 5). The toluene fluorescence signal was recorded simultaneously in two disjunct wavelength bands by a two-camera setup. After calibration, the LIF signal ratio is a proxy for the local temperature. Good agreement has been found between our new calibration curves and previously presented results by other researchers. A detailed measurement procedure is presented to minimize measurement inaccuracies and to improve precision. N-heptane was used as the base fuel and 10% of toluene was added as tracer. The toluene LIF method is capable of measuring temperatures up to 700 K; above that the signal becomes too weak. The precision of the spray temperature measurements is 4% of the temperature and the spatial resolution 1.3 mm. The fluorescence signal was also used as a fuel tracer to investigate the fuel distribution in the optical engine. . The technique was found to be very sensitive to disturbances by fluorescence of the base fuel in the wavelength range of interest. However after calibration, comparison of the temperature gradients inside the spray with Large-Eddy Simulation (LES) shows similar results. Combustion visualization Controlling ignition delay is the key to successfully enable partially premixed combustion in diesel engines. Chapter 6 presents experimental results of partially premixed combustion in an optically accessible engine, using primary reference fuels in combination with artificial exhaust gas recirculation. By varying the fuel composition and oxygen concentration, the ignition delay was adjusted. OH chemiluminescence is a useful tool to measure flame lift-off of burning fuel sprays. A similar approach has been presented using a high-speed spectral measurement setup to measure the position of the flame after injection when running in partially premixed mode. In general, increased ignition delay results in a longer lift-off length of quasi-steady flames. When combustion starts after fuel injection is completed, the longer ignition delay results in flame fronts closer to the injector due to "reflection" of the fuel against the piston wall. The mixing of fuel and air during the ignition delay period defines the local equivalence ratio. To investigate the influence of the ignition delay on the gas volume involved in combustion and the corresponding local equivalence ratio, the position of the flame is determined using high-speed visualization of OH-chemiluminescence. This enables a cycle resolved analysis of the location of OH formation, i.e. the flame position. A clear correlation was observed between ignition delay and flame location, proving that a longer ignition delay increases mixing. Emission measurements using fast-response analyzers of CO, HC and NOX confirmed the decrease in local equivalence ratio as a function of increasing ignition delay. Furthermore, multiple injection strategies were investigated, applying pilot as well as post injections, in combination with a main injection at constant load. From these results it is concluded that both pilot and post injections result in an increase of unburned hydrocarbon and CO emission and a slight decrease of nitric oxide emissions.