No description available.

No description available.

Two-stage sequential turbocharging systems are adopted in automotive powertrains, in order to enhance torque level and transient response of the internal combustion engine. High pressure (HP) and low pressure (LP) turbochargers are placed in sequence, with exhaust gasses expanding through LP and HP turbines; while, air is compressed by LP and HP compressors, respectively, before being forced into the engine intake. In order to maximise performance and flexibility of the internal combustion engine, regulating valves are implemented into the turbocharging system. In order to investigate the interaction between the boosting system and the internal combustion engines, one-dimensional (1D) model simulations are considered a fundamental tool. In this scenario, turbomachinery performances are imposed into the model through compressor and turbine maps, to analyse the complete power unit. The adoption of standard gas-stand maps would be omitting the flow non-uniformity effects on a turbomachine due to bends and the presence of heat transfer in turbochargers could influence enthalpy level at boundaries. In addition, specific to two-stage turbocharging systems, the high degree of swirling flows and temperatures at the inlet of the HP compressor induced by the upstream turbomachine can influence significantly the performance of the two-stage system. Therefore, it has been fundamental to perform this study on the advanced characterisation of turbochargers, focussing on the modelling two-stage turbocharging systems. In order to perform the research study, a new engine gas-stand facility has been developed, mapping turbochargers compressors under steady flows. In addition, in the unsteady layout, the introduction of pressure pulsations at the compressor under real engine pulsations has shown a significant shift of the surge line towards lower mass flows. In order to quantify the effects occurring in a sequential two-stage turbocharging system, an equivalent mapping approach has been developed, aiming to the improvement of accuracy in performance characterisation, accounting for inter-stage flows and heat transfers effects. Equivalent two-stage compressor maps have differed from the stand-alone mapping approach below 23Krpm due to poor accuracy of stand-alone maps at low compressor speeds. On the turbines side, maximum differences of 10% between equivalent and stand-alone maps persist in turbine efficiency, due to turbine isentropic power.

The thesis describes an investigation of the conditions which determine the starting characteristics of a synchronous motor. The asynchronous performance, at a given slip, may be estimated from the two axis operational admittances, Yd(jswₒ), Y (jswₒ), expressed as complex functions of the slip. The functions are commonly depicted as frequency response loci. Four methods, two new, are described and compared, for the experimental determination of the operational admittance loci of a synchronous motor of any size and type. The methods used for calculating the parameters from which the operational admittance loci of laminated pole synchronous motors may be derived are well known. The loci of several laminated pole synchronous motors are calculated in this manner. New two-axis equivalent circuits for the solid pole synchronous motor are derived. With the aid of a fast computer the operational admittance loci of many solid pole machines are calculated from the new equivalent circuits. The effects of flux distribution in the solid iron, saturation of the solid iron (with particular consideration of the pole tips), and the connection of adjacent solid poles by end-rings, are all considered in the calculations. Comparisons are shown between the calculated and measured operational admittance loci, obtained for both types of synchronous motors. The theoretical mean and oscillating torques and currents, derived from the loci, are then compared with the torques and currents measured. The agreement between the measured and calculated results is very good for nearly all of the machines tested, but where variations exist explanation is suggested for the discrepancy.

The computer solution to a theory of pitting crack propagation which takes into account speed effects and fluid properties is developed, and shows fair agreement with previous experimental work. The most important section of this work is the development of a technique for measuring the shape and thickness of the three dimensional oil film produced in rolling point contact. The technique uses optical interference to obtain contour maps of the oil film produced when a steel ball rolls against a glass disc which is free to rotate on an air bearing. Excellent fringes are obtained using both monochromatic and white light. Much of the improvement over earlier fringes is due to the use of a partially reflecting chrome layer on the glass disc. The optical elastohydrodynamic film thickness measurements provide a way of evaluating the relative effectiveness of different lubricant types in producing elastohydrodynamic films. The effectiveness is expresses as a relative pressure viscosity coefficient. Over 40 lubricants including many pure compounds and high performance synthetic lubricants are evaluated.. Strong evidence for the non-Newtonian behaviour of polymer blended lubricants, and some evidence for the viscoelastic behaviour of.pure compounds is found. The contour map of the oil film provides an excellent starting point for the theoretical investigation of the elastohydrodynamic problem in point contact. A computer program for calculating the pressure field from the film thickness contour, using Reynolds' equation, is developed. Less success is achieved in the elasticity part of the problem. Finally, the further possibilities for developing both the experimental and theoretical techniques are discussed.

A treatment has been developed for the dynamic analysis of single and double span guyed masts acted upon by harmonic loading along their lengths. Solutions for forced vibrations of a mass on a non-linear spring, with or without linear viscous damping and subjected to harmonic excitation have been developed first. These solutions emphasize the importance of higher order harmonic.terms occurring in the response. The properties of the fundamental and the higher order harmonics have been studied. Next, general static analysis methods for a single guy have been developed. These have been extended to the treatment of in-plane vibration under the influence of forced in-plane vibration of the supports. The occurrence of higher order harmonics in the guy motion and the guy tension has been established. Next, methods have been developed for solving the guyed mast problem so that the response of the structure to harmonic excitation along the mast can be obtained by simultaneous solution of the equations of motion for the guy attachment points. The dynamic response of the mast proper, idealised as an elastic beam, has been incorporated in these equations by using suitable coefficients. Experimental investigations on the model of a single guy with forced vibration of the upper support and that of a one span mast arranged as a single degree of freedom system have been made. The agreement between the theoretical and the experimental results has been found to be satisfactory. The exact theoretical solutions for the one span and two span masts have been compared with various approximate solutions. The importance of higher order harmonics in the resulting motion has been demonstrated. The methods developed are completely general so that they can be extended to take into account aspects like wind loading on the guys, transverse vibrations of the guys etc.

This thesis examines the steady state power system problem and introduces the author's investigations directed towards producing analysis methods which Can be applied to a number of problems, such as the load flow, minimum loss and economic dispatching problems, for which no unified approach has been available in the past. Load flow techniques are summarized and new iterative methods are introduced based on a mesh description of the system and on a hybrid admittance—impedance formulation. Criteria for the selection of reference and slack busses are presented. Nonlinear programming techniques are discussed and are shown to provide a feasible mathematical context for the solution of not only the load flow problem but also the optimization problems involving minimum transmission losses and economic dispatching. To extend the scope of nonlinear programming methods to large systems, decomposition techniques are developed and applied success— fully to a number of systems. Numerical results of actual systems are provided for all new methods introduced in this thesis,

This thesis on resistance to flow in open channels includes an experimental investigation of some of the basic actions and details of the steady and of the unsteady parts of the motion. The experiments were done in a glass walled channel using clear water and with smooth as well as rough beds of non-mobile material though one bed was of ripple mark type having been generated hydraulically and subsequently consolidated. Certain analytical expressions for the friction coefficient for laminar flow in wide rectangular section are verified experimentally and the lower value of the critical Reynolds number for this section is established. To find the effect of the air water interface on the turbulence, measurements are made in two-dimensional fully developed turbulent flows using a photographic technique and the results are compared with those previously obtained in pipes. Also the effect of roughness on turbulence is investigated and the results with different roughnesses are correlated. In an attempt to detect the secondary currents in straight channels, a photographic technique is developed to visualize and record the motion across the main flow. From the photographs some patterns of the secondary currents are mapped out. iii The problem of resistance to flow is finally considered. Smooth channel results are compared with those of smooth pipes and the effect of depth width ratio on the resistance is disclosed. For the channel with various forms of bed roughness and smooth side walls, the surface roughness in terms of equivalent sand roughness is evaluated from velocity distributions and from the friction coefficient results. It is shown that for such cases the method, proposed by Einstein and Johnson, commonly used to eliminate the effect of side walls for evaluating the bed roughness gives misleading results..

When a fluid flows through a curved pipe, a secondary flow is set up, the fluid in the middle of the pipe moving outwards and that near the walls moving inwards. Near the entrance to the bend, the secondary flow changes a great deal and in some cases appears oscillatory. If the bend is of sufficient length, the changes in secondary flow are damped by viscosity and a region where the flow is fully developed is reached. It has been shown theoretically and experimentally, that the dynamical similarity of fully developed laminar flow depends on a non-dimensional parameter K/W = 2aWo/ν (a/R) 1/2, where Wo is the mean axial velocity, ν is the coefficient of viscosity and 'a' is the radius of the pipe which is bent into a circle of radius R.

No description available.