Knock In Spark Ignition Engines Research Articles

Biphasic sensitization effect of NO2 on n-C4H10 auto-ignition

NOx in the exhaust gas can influence the auto-ignition properties, thus are important for the design and operation of HCCI engines and predicting engine knock in spark ignition engines. More work is clearly required to improve the understanding on fuel/NOx chemistry for both reaction mechanism construction and engine application. In this study, n-C4H10, as a representative hydrocarbon fuel with the negative temperature coefficient (NTC) behavior, was selected for test. Ignition delay times of n-C4H10/air mixtures with varying NO2 addition (0, 500 ppm, 1%, and 5%) have been measured behind reflected shock waves. The sensitization effect of NO2 on n-C4H10 auto-ignition was investigated at engine relevant conditions covering temperatures from 700 to 1200 K, pressure of 20 atm, and equivalence ratios from 1.0 to 2.0. Besides, the mixtures with excessive amounts of NO2 (5% and 10%) were also studied at relatively high temperatures and 10 atm, aiming at highlighting the interactions between NO2 and n-C4H10. Results indicated the effect of NO2 on n-C4H10 oxidation exhibits prominent temperature and NO2 concentration dependences. Trace NO2 addition (500 ppm) promotes low-temperature reactivity of n-C4H10 and reduces the ignition delay times. While higher levels of NO2 no longer promote ignition, instead, it turns gradually to be an inhibiting effect. At high temperatures, NO2 dramatically accelerates the n-C4H10 oxidation, and the promoting effect peaks at around 5% NO2 then saturates with further NO2 addition. In the negative temperature coefficient region, the NTC behavior of n-C4H10 is weakened and even disappears by the introduction of NO2. A preliminary detailed kinetic model was proposed to describe the n-C4H10/NO2 chemistry, and in general, is capable of capturing the NO2 sensitization behaviors under different conditions. Kinetic analyses were then performed to clarify the sensitization mechanisms of NO2 and the chemical interactions between NO2 and n-C4H10.

End-Gas Autoignition Mechanism in a Downsized Spark-Ignition Engine: Effect of Inhomogeneity

ABSTRACT It is generally accepted that knock in spark-ignition engines is caused by end-gas autoignition. However, limited studies regarding to the effect of the inhomogeneity of temperature and composition on end-gas autoignition are presented. The major objective of this paper is to quantitatively understand the flame propagation velocity and the end-gas autoignition mechanism affected by the reactivity gradient formed by thermal stratification and composition stratification in the combustion chamber of a downsized SI engine. The conclusions were obtained based on a knock combustion case under an initial condition of high temperature and high pressure. In the present work, fuel stratification has a significant influence on the propagation velocity of the main flame. The flame propagation velocity declined with an increase in the level of fuel stratification. However, the temperature stratification had little effect on the flame propagation velocity. It was found that the onset of autoignition advanced with increasing the level of temperature stratification, which demonstrated an approximately linear relationship. It can be found that a high level of fuel stratification induced deflagration because of multi-spot spontaneous combustion. On the contrary, a very intense knock with detonation wave was observed under a slight level of temperature stratification. Under this condition, the temperature gradient near the nascent AI kernel was small and contributed to sequential spontaneous combustion. In fact, both the fuel stratification and temperature stratification are essentially the reactivity stratification. Therefore, similar results also can be conducted from fuel stratification. This work will give a new insight into suppressing knock mechanisms by different strategies.

Laminar Flame Speeds of Premixed Iso-Octane/Air Flames at High Temperatures with CO <sub>2</sub> Dilution

<div class="section abstract"><div class="htmlview paragraph">Spherically expanding flames are employed to measure the laminar flame speed of premixed iso-octane/air mixtures at elevated temperatures through both experiments and numerical simulations. Iso-octane (2,2,4-trimethlypentane) is an important gasoline primary reference fuel (PRF). While most studies on laminar burning velocity of iso-octane focus on low temperatures (less than 400 K), the experiments here were conducted in an optically accessible constant volume combustion chamber between 373 K-473 K, at a pressure of 1 bar, and from ϕ=0.8 to ϕ=1.6. The effect of diluent is investigated through the addition of 15% CO<sub>2</sub> dilution in order to simulate the effect of exhaust gas recirculation. The decreased reactivity with diluent addition reduces mixture reactivity, which can reduce the propensity for knock in spark ignition engines. All laminar flame speeds were calculated using the constant pressure method enabled via schlieren visualization of the spherically propagating flame front. Results show that laminar flame speeds of iso-octane/air mixtures at 1 bar increase by 46-51% with initial temperature increases from 373 K to 473 K. However, the addition of 15% CO<sub>2</sub> dilution to the iso-octane/air mixtures at 1 bar and 473 K results in a 47-51% decrease in the laminar burning velocity. Numerical results obtained with CHEMKIN [<span class="xref">1</span>] using the kinetic mechanism of Chaos et al. [<span class="xref">2</span>] show excellent agreement with experimental data for ϕ=0.8-1.2; however, the kinetic mechanism slightly underestimates the experimentally observed laminar flame speeds between ϕ=1.3-1.6.</div></div>

Kinetics of autoignition: a simple intuitive interpretation and its relation to the Livengood-Wu integral.

It is well known that the gas-phase autoignition phenomenon often involves branched chain reactions as well as the acceleration of reactions by thermal feedback. Despite the huge combustion kinetic mechanisms of large hydrocarbons found in practical fuels, chain reactions in the early stages of alkane autoignition exhibit simple kinetics since the pseudo-first-order assumption and the linear approximation are valid. In this study, this simple picture of autoignition will be presented starting from the H2-O2 system and then extending to practical fuel-air mixtures. The present interpretation gives the theoretical rationale for the Livengood-Wu integral which is known as an empirical method to predict the timing of knock in spark-ignition engines.

Experimental investigation on knocking combustion characteristics of gasoline compression ignition engine

Gasoline compression ignition (GCI) is one of the most promising combustion concepts to maintain low pollutant emissions and high efficiency. However, knock phenomenon is an obstacle to GCI combustion. In this paper, experimental investigation is conducted to study GCI knocking characteristics. Firstly, analysis of engine performance and knocking trends was conducted, which shows an increasing trend of both brake mean effective pressure (BMEP) and knock intensity as injection timing advances. Secondly, comparison of knocking characteristics between the modes of spark-ignition (SI) and GCI was performed. The results of statistical and heat release analysis show that the combustion process of the two combustion modes are quite different. Knock in SI engine is a random phenomenon caused by auto-ignition of end gas. On the contrast, knock in GCI engine is ascribed to local rapid burning rate, which doesn't happens randomly. Finally, double-injection strategy is applied to obtain different mixture in premixed degree. With more fuel pilot injecting into cylinder, higher BMEP and knock intensity can be observed. The results shall give insights into the possibilities of getting better engine performance while controlling cycle-to-cycle variations and knock intensity in acceptable regions.

Quantification of knock benefits from reformate and cooled exhaust gas recirculation using a Livengood–Wu approach with detailed chemical kinetics

In this work, a simple methodology was implemented to predict the onset of knock in spark-ignition engines and quantify the benefits of two practical knock mitigation strategies: cooled exhaust gas recirculation and syngas blending. Based on the results of this study, both cooled exhaust gas recirculation and the presence of syngas constituents in the end-gas substantially improved the knock-limited compression ratio of the engine. At constant load, 25% exhaust gas recirculation increased the knock-limited compression ratio from 9.0 to 10.8:1 (0.07 compression ratio per 1% exhaust gas recirculation) due to lower end-gas temperature and reactant (fuel and oxygen) concentrations. At exhaust gas recirculation rates above 43%, higher intake temperature outweighed the benefits of lower end-gas reactant concentration. At constant intake temperature, cooled exhaust gas recirculation was significantly more effective at all exhaust gas recirculation rates (0.10 compression ratio per 1% exhaust gas recirculation), and no diminishing returns or optimum was observed. Both hydrogen and carbon monoxide were also predicted to improve knock by reducing end-gas reactivity, likely through the conversion of high-reactivity hydroxy-radicals to less reactive peroxy-radicals. Hydrogen increased the knock-limited compression ratio by 1.1 per volume percent added at constant energy content. Carbon monoxide was less effective, increasing the knock-limited compression ratio by 0.38 per volume percent added. Combining 25% cooled exhaust gas recirculation with reformate produced from rich combustion at an equivalence ratio of 1.3 resulted in a predicted increase in the knock-limited compression ratio of 3.5, which agreed well with the published experimental engine data. The results show the extent to which syngas blending and cooled exhaust gas recirculation each contribute separately to knock mitigation and demonstrate that both can be effective knock mitigation strategies. Together, these solutions have the potential to increase the compression ratio and efficiency of spark-ignition engines.

Thermal and Ludwig–Soret diffusion effects on near-boundary ignition behavior of reacting mixtures

The autoignition behavior is key towards providing insight into a fuel's behavior and performance and equally important towards developing kinetic models. However, there has been evidence in past literature that under certain conditions legacy reactor experiments may exhibit inhomogeneities that could potentially reduce the scientific value of the reported data. Additionally, the initiation of knock in spark ignition engines has been observed to occur close to the cylinder wall. In this study, detailed one-dimensional simulations were carried out in order to provide additional insight into the aforementioned observations. The main focus of the present investigation is on the effects of a colder wall relatively to the core of the reacting mixture, and the attendant development of thermal boundary layers. It is determined that the thermal stratification could alter the ignition behavior for fuels that exhibit distinct negative temperature coefficient behavior and could cause also light and heavy species concentration stratification due to Ludwig–Soret diffusion. More specifically, simulations performed for n-heptane/air mixtures revealed that localized exothermic centers could develop for a range of initial mixture temperatures. Furthermore, simulations for hydrogen/oxygen/argon mixtures showed that species stratification caused by Ludwig–Soret diffusion could lead to increased local heat release rates.

A RANS-Based CFD Model to Predict the Statistical Occurrence of Knock in Spark-Ignition Engines

A RANS-Based CFD Model to Predict the Statistical Occurrence of Knock in Spark-Ignition Engines

End-gas autoignition and detonation development in a closed chamber

It is generally accepted that knock in spark ignition engines might be caused by end-gas autoignition. However, the detailed mechanism for autoignition-induced pressure oscillation and detonation development is still not well understood. This work studied end-gas autoignition and detonation development in a closed chamber using 1D simulation. Stoichiometric hydrogen/air mixture at different initial temperatures and pressures was considered and detailed chemistry was included in simulation. The objectives were to identify possible modes of end-gas combustion and to understand the mechanism of autoignition-induced pressure wave and detonation development. Depending on the chamber length as well as the initial temperature and pressure, there are three modes of end-gas combustion: normal flame propagation without autoignition, autoignition without detonation development, and detonation development. The amplitude of pressure oscillation was found to be determined by the mode of end-gas autoignition: autoignition can induce high amplitude of pressure oscillation similar to conventional knock; and detonation development can cause extremely high amplitude of pressure oscillation similar to super-knock. It was shown that autoignition and detonation development can be induced by increasing the initial temperature, initial pressure, or chamber length. The evolution of states of different flow particles was tracked and the combustion mode was found to switch from constant-pressure to constant-volume when autoignition occurs. The coupling between pressure wave and chemical reaction was analyzed and the mechanism for autoignition front acceleration and detonation development was investigated. Moreover, autoignition in end-gas with different values of ignition progress was simulated. It was demonstrated that high reactivity of end-gas promotes autoignition and detonation development.

Étude expérimentale et modélisation des réactions d’oxydation du néopentane et de l’isopentane

The different behaviour of hydrocarbons with respect to autoignition phenomena is, at present, not yet fully explained. We have therefore investigated the oxidation of two alkanes that have different octane numbers: neopentane (85.5) and isopentane (92.3), to obtain a better understanding of their reaction mechanisms, in particular, those reactions that are responsible for the onset of knock in spark ignition engines. The experimental study was performed at 873 K in a jet-stirred flow reaction vessel. The oxidation mechanisms were simplified by using the CHEMKIN programme of simulation of reaction mechanisms. These mechanisms were compared to those accounting for the oxidation of n-pentane, cyclopentane, n-heptane, and isooctane that we have previously studied. This comparison shows that the different behaviour of these hydrocarbons can be explained, at least in part, by the presence, in the reaction medium, of resonance-stabilized radicals.Key words: oxidation, neopentane, isopentane, autoignition, modelling.

A diagnostic bi-zonal combustion model for the study of knock in spark-ignition engines

A bi-zonal model for the combustion process in spark-ignition engines was developed to provide thermodynamic information, using experimental pressure records. The model was applied to the knock analysis, introducing a reduced kinetic mechanism. The unburned gas temperature in an adiabatically compressed core region is shown to be a good approximation for the reaction temperature. The activation energy of the forward rate of the isomerization reaction was determined for each of seven fuels by comparison with the experiments. It is shown that the activation energy has an almost linear dependence on the research octane number.

Étude expérimentale et modélisation des réactions d'oxydation du n-pentane et du cyclopentane

The reduction of automotive emissions and the reformulation of gasoline composition require a better understanding of the combustion mechanism, in particular aiming at the determination of the influential reactions which are responsible for the onset of knock in spark ignition engines. We have studied the oxidation mechanism of two hydrocarbons of different research octane number: n-pentane (62) and cyclopentane (100). This work follows a previous study on the oxidation of n-heptane and isooctane. The experimental study was performed at 873 K in a jet-stirred flow reaction vessel. The reaction mechanisms of n-pentane and cyclopentane were analysed on a purely kinetic basis, derived from our experimental results, and were finally modelled by using a programme of simulation of reaction mechanisms. The different behaviour of these two hydrocarbons with respect to autoignition phenomena is mainly dependent on the relative concentration of resonance-stabilized radicals in the reaction medium. Keywords: oxidation reaction, n-pentane, cyclopentane, modelling, autoignition.

Cars temperature measurements and the cyclic dispersion of knock in spark ignition engines

The use is described of the coherent anti-Stokes Raman spectroscopy (CARS) technique in a single-cylinderresearch engine to measure end-gas temperature in the regime that leads to the onset of knock. The investigations, under nominally the same running conditions, covered a large cyclic dispersion of knock, with knock occurring in about 80% of the cycles. In-cylinder pressures were measured with a transducer, and an ion gap signified flame arrival towards the end of the flame travel and just before autoignition. In general, the temperature increased by over 100 K, both with and without knock, as a result of exothermic preflame reactions. This measured elevation was supported by computations with a simple five-step model of autoignition, calibrated against measured autoignition times in a rapid compression machine. The temperatures measured just prior to knock increased with the severity of the knock. The observed severe cyclic dispersion of knock, under the conditions of the study, is attributed primarily to the cyclic dispersion of flame speeds. An increase in flame speed is associated with an increase in temperature and greater propensity to knock.

Novel features of end-gas autoignition revealed by computational fluid dynamics

Autoignition of the end-gas in the combustion chamber is believed to be a primary cause of knock in spark ignition engines. Fundamental studies of the autoignition of hydrocarbon + air mixtures has been performed successfully in rapid compression machines. There has also been considerable progress towards the numerical modeling of these phenomena, but applications to combustion in a rapid compression machine have been based only on a spatially uniform well-mixed condition. The authors report in this paper the implementation of a fluid dynamics code in conjunction with the Shell generalized (or reduced) kinetic model, which represents the exothermic oxidation of the alkane components of gasoline, to predict the development of autoignition in the combustion chamber. The numerical simulations reported in this paper were matched to the performance of the rapid compression apparatus in use at Leeds. In this system, combustion occurred in a cylindrical chamber, the flat, end faces of which were the piston crown and cylinder head. The reactants were compressed by a piston, which was then brought to rest at the end of its stroke. The geometric configuration of the combustion chamber of the rapid compression machine is fully axisymmetric. This permitted a two-dimensional treatment of spatial conditionsmore » on a plane representing piston displacement and half radius. Reactive gas flows were simulated on a 14 [times] 50 mesh at the start of compression, which was reduced to 14 [times] 10 cells as the piston reached top-dead-center. The resolution within the vicinity of the cylinder wall was enhanced by a nonequidistant grid structure.« less

An analytical approach to auto‐ignition and knock in internal combustion engines

A simplified analytical approach to the problem of determining the conditions for the onset of knock in spark ignition engines is made, using a criterion for knock based on a hypothetical gross chemical reaction of the fuel-air mixture in the end gas. It is then applied to auto-ignition studies in motored engines. Parameters such as compression ratio, initial temperature and pressure, fuel properties, ignition timing, speed and mixture strength are considered. This simplified treatment may form a basis for interpolating data, particularly when few parameters are being varied at a time. Some aspects of the treatment are illustrated by experimental engine data, drawing whenever possible from already published results.

Some observations by schileren technique of the propagation of flames in a closed vessel

The study of flame propagation at constant volume presents problems of particular technical interest both in the analysis of flame propagation and in the application of the results to such practical phenomena as knock in spark ignition engines. In recent years various photographic combustion studies in engine-like apparatus have been made to elucidate flame propagation under actual engine conditions, including the mechanism of knock. In this field the work of Miller (I), Osterstrom (2), Taylor, Taylor, et al. (3), is outstanding. The course of the rapid reactions taking place in the engine cylinder has been followed by stroboscopic sampling by Egerton, et al. (4), and more recently by Downs, et al. (5). As a result of these various experimental studies, it now appears possible to present a reasoned explanation of the knock process at least for some fuels. Precise analysis of flame propagation in the above studies is made difficult by the size and shape of the combustion chamber and also by piston and port induced gas movements. In the present experiments combustion processes in a long, constant volume bomb were studied using simultaneously direct, and Schlieren photography. Preliminary investigation has provided records showing in fine detail, the form and structure of the flame front at all stages of the combustion process and has enabled a study to be made of such phenomena as flame arrest, and the development of pre-reactions in the end gas leading to autoignition and knock.