Alexey Burluka

Ultra Boost for Economy: Extending the Limits of Extreme Engine Downsizing

The paper discusses the concept, design and final results from the ‘Ultra Boost for Economy’ collaborative project, which was part-funded by the Technology Strategy Board, the UKs innovation agency. The project comprised industry- and academia-wide expertise to demonstrate that it is possible to reduce engine capacity by 60% and still achieve the torque curve of a modern, large-capacity naturally-aspirated engine, while encompassing the attributes necessary to employ such a concept in premium vehicles. In addition to achieving the torque curve of the Jaguar Land Rover naturally-aspirated 5.0 litre V8 engine (which included generating 25 bar BMEP at 1000 rpm), the main project target was to show that such a downsized engine could, in itself, provide a major proportion of a route towards a 35% reduction in vehicle tailpipe CO2 on the New European Drive Cycle, together with some vehicle-based modifications and the assumption of stop-start technology being used instead of hybridization. In order to do this vehicle modelling was employed to set part-load operating points representative of a target vehicle and to provide weighting factors for those points. The engine was sized by using the fuel consumption improvement targets and a series of specification steps designed to ensure that the required full-load performance and driveability could be achieved. The engine was designed in parallel with 1-D modelling which helped to combine the various technology packages of the project, including the specification of an advanced charging system and the provision of the necessary variability in the valvetrain system. An advanced intake port was designed in order to ensure the necessary flow rate and the charge motion to provide fuel mixing and help suppress knock, and was subjected to a full transient CFD analysis. A new engine management system was provided which necessarily had to be capable of controlling many functions, including a supercharger engagement clutch and full bypass system, direct injection system, port-fuel injection system, separately-switchable cam profiles for the intake and exhaust valves and wide-range fast-acting camshaft phasing devices. Testing of the engine was split into two phases. The first usied a test bed Combustion Air Handling Unit to enable development of the combustion system without the complication of a new charging system being fitted to the engine. To set boundary conditions during this part of the programme, heavy reliance was placed on the 1-D simulation. The second phase tested the full engine. The ramifications of realizing the engine design from a V8 basis in terms of residual friction versus the fuel consumption results achieved are also discussed. The final improvement in vehicle fuel economy is demonstrated using a proprietary fuel consumption code, and is presented for the New European Drive Cycle, the FTP-75 cycle and a 120 km/h (75 mph) cruise condition.

The influence of simulated residual and NO concentrations on knock onset for PRFs and gasolines

Modern engine developments result in very different gas pressure-temperature histories to those in RON/MON determination tests and strain the usefulness of those knock scales and their applicability in SI engine knock and HCCI autoignition onset models. In practice, autoignition times are complex functions of fuel chemistry and burning velocity (which affects pressuretemperature history), residual gas concentration and content of species such as NO. As a result, autoignition expressions prove inadequate for engine conditions straying far from those under which they were derived. The currently reported study was designed to separate some of these effects. Experimental pressure crankangle histories were derived for an engine operated in skip-fire mode to eliminate residuals. The unburned temperature history was derived for each cycle and was used with a number of autoignition/knock models. A simple empirical expression proved no less effective than more complex formulations in predicting knock onset for iso-octane and PRFs over a wide range of residual free operating conditions. Prediction of knock onset for two commercial gasoline fuels proved less reliable, but was improved using an octane index correction method. Computations of knock onset times proved sensitive to simulated residual gas/EGR and NO concentrations. The influence of NO proved variable and contrary for iso-octane, gasolines, primary and toluene reference fuel mixtures.

Burn Rate Implications of Alternative Knock Reduction Strategies for Turbocharged SI Engines

This work is concerned with the analysis of different charge dilution strategies employed with the intention of inhibiting knock in a high output turbocharged gasoline engine. The dilution approaches considered include excess fuel, excess air and cooled external exhaust gas re-circulation (stoichiometric fuelling). Analysis was performed using a quasi-dimensional combustion model which was implemented in GT-Power as a user-defined routine. This model has been developed to provide a means of correctly predicting trends in engine performance over a range of operating conditions and providing insight into the combustion phenomena controlling these trends. From the modelling and experimental data presented, it would appear that the use of cooled externally re-circulated exhaust gases allowed fuel savings near to those achieved via excess air, but with improved combustion stability and combustion phasing closer to the optimum position.

Effect of Supercharging on Cycle-To-Cycle Variation in a Two-Stroke Spark Ignition Engine

Fluctuations in the operational output of spark ignition engines are observed from one engine cycle to the other, when an engine is run at technically identical operating condition. These fluctuations known as cycle-to-cycle variations, when high, adversely affect the performance of an engine. Reduction in cycle-to-cycle variation in engines has been noted by researchers as one of the methods of improving engine efficiency and operational stability. This study investigated the combustion performance characteristics of two fuels: E5 (95% gasoline and 5% ethanol) and ULG98 (unleaded gasoline) in a spark ignition engine, operating at varying inlet pressure conditions and ignition timing. A two-stroke, 80mm bore, spark ignition engine was operated at an engine speed of 750 rpm, inlet pressures of 1.6 and 2.0 bar and spark-timings ranging from 2 to 13 bTDC. A top cylinder head with a centralized spark plug was used in all the experiments. The Gross Indicated Mean Effective Pressure (GIMEP), GIMEP’s coefficient of variance (COV) and the Mass Fraction Burned (MFB) were determined. An increase in the intake pressure of the engine gave rise to an increase in the engine’s GIMEP in all the fuels tested. While a decrease in the COV of the engine’s GIMEP with an increase in the engine’s intake pressure was observed, its effect on the crank angle of occurrence of MFB varied for E5 and ULG 98 fuels. The study established that reduction in the cycle-to-cycle variation in spark ignition engines can be achieved by supercharging or turbo-charging these engines.

An Investigation of Various Chemical Kinetic Models for the Prediction of Autoignition in HCCI Engine

Diverse kinetic models for iso-octane, n-heptane, toluene and ethanol i.e. main gasoline surrogates, have been investigated. The models have different levels of complexity and strong and weak points. Firstly, ignition delay times for various fuel blends have been calculated and compared with published shock tube measurements. Kinetic models which are capable of distinguishing between Primary and Toluene Reference Fuels have been used further on in a zero-dimensional Homogeneous Charge Compression Ignition engine model to predict auto-ignition. The modelling results have been compared to the experimental results obtained in a single cylinder research engine. A discussion has been made on the ability of these models to predict autoignition.Copyright

Average Vaporisation Rate in Turbulent Subcritical Two-Phase Flow

This work considers alternative expressions for turbulent evaporation rate used in the framework of an entirely Eulerian model based on a transport equation for the average liquid surface area. Commonly employed expressions for vaporisation rate derived from Spalding-Godsave theory fail to account for vaporisation enhancement induced by turbulence; moreover, they do not describe experimentally observed fact that the pressure affects vaporisation rate differently in a turbulent and a laminar flow. To address these shortcomings, an alternative formula for the vaporisation rate is proposed based on an assumption that the vaporisation rate is governed by a small-scale turbulence. This model is assessed for a range of pressure and temperature conditions, using experiments of Brandt et al. (1997a) performed in a premix duct with a flat-bed atomiser as the test case. Turbulence intensities and scales in the chosen test case are typical for a modern gas-turbine combustion chamber. This new expression results in prediction of evaporation rate and SMD in a good agreement with experimental results.

The effect of methylfurans on the physicochemical and performance characteristics of finished motor gasoline

The physicochemical and performance characteristics of blends of commercial gasoline with the promising oxygenate antiknock additives 2-methylfuran and 2,5-dimethylfuran have been studied. The key parameters determining the compliance of gasoline with the GOST R 51866-2002 (EN 228-2004) requirements have been measured. A high antiknock activity of the test compounds has been demonstrated, and blending octane numbers have been calculated. It has also been found that the addition of 2-methylfuran and 2,5-dimethylfuran to gasoline substantially deteriorates its oxidation stability and increases gum content.

Combustion and Autoignition Modelling in a Turbocharged SI Engine

A holistic modelling approach has been employed to predict combustion, cyclic variability and knock propensity of a turbocharged downsized SI engine fuelled with gasoline. A quasi-dimensional, thermodynamic combustion modelling approach has been coupled with chemical kinetics modelling of autoignition using reduced mechanisms for realistic gasoline surrogates. The quasi-dimensional approach allows a fast and appreciably accurate prediction of the effects of operating conditions on the burn-rate and makes it possible to evaluate engine performance. It has also provided an insight into the nature of the turbulent flame as the boost pressure and speed is varied. In order to assess the sensitivity of the end-gas chemical kinetics to cyclic variability, the in-cylinder turbulence and charge composition were perturbed according to a Gaussian distribution. Coupling cyclic variability with autoignition modelling allowed prediction of the autoignition propensity for the entire spectrum of cyclic variations in cylinder pressure. The models have been validated against engine test data from a technology demonstrator downsized, turbocharged engine. The knock-limited spark advance was predicted for a RON 95 and RON 102 gasoline within 2° of crank angle. This work demonstrates the viability of chemical kinetics for gasoline surrogates coupled with 0-dimensional thermodynamic modelling approach as a fast and reliable development tool for high performance engines.

Modelling of a Turbulent Jet in a Gas Crossflow

The first part of this work presents a comparison of predictions obtained with several two-equation type RANS turbulence models commonly used in industry against experimental data obtained by Whitelaw et al [1]. All examined models yield a relatively poor match in the flow region very close to the wall; agreement with the measurements improves significantly when moving further away from the wall. This concerns both the internal normal stress profiles and the average velocity profiles, the latter show improved prediction of the recirculation zone area when moving further into the main stream. Downstream behaviour for both models shows an excellent match more than 6 diameters away from the jet inlet, defined as the region after which the flow essentially resumes its normal duct behaviour[1].Expanding upon these RANS results, another series of simulations using LES modelling with the standard Smagorinsky SGS model was conducted using the same grid and compared to the RANS-based results. Although performance in the most complex flow areas was slightly improved over RANS, this was at the cost of an increase of computation time by almost a factor of 6.The next stage involved developing a code based on the model for two-phase flow described in [2] to predict the atomisation pattern for a non-vaporising (or “cold”) flow based on the parameters of the previous simulations. This model implements transport equations for the liquid mass fraction and the average surface area per unit mass along with an equation for average density; resulting in an entirely Eulerian model which can be used to predict atomisation from first principles. Current work consists in development of additional source terms describing vaporisation in a strongly turbulent environment and further coupling with a combustion model applicable to the combustion chamber of an industrial gas turbine.Copyright

Computational fluid dynamic modelling of atomisation processes in turbulent flows

Abstract Atomisation of liquids is frequently encountered in the liquid-gas flows used in many practical chemical and process engineering applications, and an ability to reliably predict such flows is of benefit to the optimisation and performance improvement of existing equipment and processes, as well as the evaluation of retrofit options and the design of new equipment, systems and plant. This paper considers the ability of an Eulerian, two-equation continuum model of the atomisation process, embodied within a computational fluid dynamic framework, to reproduce the experimentally established behaviour of air-assisted atomisation. The influence of injector exit velocity profile, surface tension, gas velocity, and liquid and gas densities on predictions of the model is examined, and results found to be in good agreement with available experimental data.