Combustion of Alternative Fuels: Opto-Mechanical Design and Optical Investigation

Abstract

There is a considerable body of work based on observations and measurements obtained from optically accessible high pressure reactors, such as piston engines and constant volume combustion chambers. However, there are still important aspects of design and relevant information regarding optical access that are missing or are insufficiently explored or not readily accessible in the existing literature. A comprehensive review of requirements for optical access to such high-pressure, high-temperature systems has been conducted. It is presented in a readily navigable format, with data and technical correlations hitherto not found in a ‘user-friendly’ style, along with a comparison of optical materials, relevant properties and guidelines for application. It was found that optical materials often have sufficient mechanical strength, but that there were limiting factors of design, namely: working temperature, required electro-magnetic (EM) wave range and cyclic loading. It is especially difficult to carry out accurate design with cyclic loading as in a running engine, as the relevant data for optical materials is lacking in the literature. As a result of the significant uncertainty arising from inconsistency of design data, it is found that a safety factor is required, between two and five, depending on the probability of failure and associated risk. The opto-mechanical design process of pressure vessels is shown in detail for two real case- studies: a high-pressure steady-flow combustor Abstract v and the combustion chamber of a spark ignition engine. The latter system is then adopted and utilised for further diagnostic analysis. Utilising the optical access, a high-speed imaging system was used to record the ignition and flame kernel formation in the internal combustion engine. A range of fuels were investigated, including gasoline, isooctane, and a few alternative renewable fuels: E85, M85, and hydrogen. The experiments were conducted at stoichiometric conditions for the liquid fuels and ��=0.67 for hydrogen at various engine speeds and compression ratios. A novel analysing method was proposed to process the acquired raw optical data, where ellipses, rather than conventional spheres, were fitted onto image projections of the visible light emitted by the flames. A cross-comparison of the results with the available data from the literature was also conducted. The large amount of optical data allowed the statistical evaluation of the flame area, flame speed and a flame-shape descriptor. The image analysis showed that the ellipse fitting method provided a 50–100 % better fit and thus allowed a more accurate description of the flame propagation properties. The results indicated that gasoline and isooctane had similar flame propagation behaviour, but a significant difference was observed between these fuels and E85, M85 and hydrogen. Similarities were found between the propagation characteristics of M85 and hydrogen, showing the fastest Abstract vi propagating flames among all the fuels. The statistical analysis found that the precision of the flame speed measurement and the roundness of the flames increase with the engine speed, compression ratio, and time elapsed after ignition.