Prediction of heat flow, temperature and stress in diesel engine pistons incorporating thermal barriers.

Abstract

As part of a broader research project of adiabatic engine investigation, the present work concerns the effect of heat barriers in the suppression of heat loss and on the thermal loading conditions in pistons, from both a steady-state and transient point of view. The theoretical work has been concerned primarily with the development of a comprehensive set of computer programs dealing with either the steady-state axisymmetric heat transfer and stress analysis, or the transient heat transfer analysis. In developing the programs, the present work provides a subdomain integral approach to the finite element method, the integration of which can be either weighted or non-weighted. In the latter case, the subdomain integral can often be physically interpreted as, e.g. the energy balance in a control volume or the force equilibrium on a free body, hence it provides much more intuitive and physical insights to the finite element method, with the same accuracy and same applicability to solving various engineering differential equations as the Galerkin method. The subdomain integral approach can be directly applied to the initial value problems, where the subdomain is offset in the time-dimension. By this method the time-marching process can be represented in the form of a system matrix, which enables the direct solution of periodic phenomena without the requirement of incorporating and iterating the guessed initial conditions from cycle to cycle. In this way, the convergence from cycle to cyle is inherently satisfied. More importantly, the present work also provides a one-dimensional linearization model of the axisymmetric heat transfer and thermal stress field, from which may be deduced the equivalent thermal resistance (two-terminal model) or its three components (three-terminal model) of the configuration, as well as the thermal stress sensitivity to the thermal loadings. By incorporating the equivalent thermal resistances into the engine cycle simulation program, the heat flow rate over the full cycle in the engine cycle simulation can be matched with the heat flow rate through the combustion chamber components. The experimental work has concentrated on methods for determining the gas-side boundary conditions from measured piston temperatures. The area-mean wall temperatures (and hence the total heat flow through the piston when the effective gas temperature is known), can be obtained from the measured maximum surface temperatures by applying the present one-dimensional model. Furthermore, a novel differential probe is proposed to measure the heat transfer coefficient and the effective gas temperature simultaneously. By applying these specially developed finite-element programs, matched either with the experimentally determined boundary conditions or with the corresponding engine cycle simulation program, a detailed investigation on various heat barrier configurations, concentrating on a specially designed and manufactured air-gap piston, leads to the following conclusions: 1. The practical range of insulation design lies between 40-65% reduction of heat loss. The air-gap piston design presented in this thesis can give about 50% reduction of heat loss, which is in the desired range of insulation. 2. The effect of insulation on the percentage suppression of heat loss is influenced by the engine rating conditions, and by the gas-side boundary conditions. The higher the rating, or the higher the gas-side heat transfer coefficient, the more significant is the effect of insulation on the suppression of heat loss. 3. In the "ideal transient adiabatic" operation, about 50% of the total recovered heat can be converted into work. However, in real thermal barrier configurations, the wall surface temperature fluctuation is small. Therefore, nearly all real thermal barriers (including the air-gap design) undergo "steady-state" operation, where only about 30% of the total recovered heat can be converted into work. 4. Insulation of the piston is not only tolerable, but actually favourable to piston ring life as the ring groove temperature is well below the permissible range. The maximum radial thermal expansion in a high temperature air-gap piston is well below the cold tolerance limit between the piston and the liner. 5. The fatigue analysis shows that the present air-gap design satisfies the demands of fatigue strength. The cyclic thermal stresses in the air-gap piston do not add significantly to the compressive stresses on the crown. 6. By matching the finite element analysis with the engine cycle simulation program, the reciprocal effect of insulation on the gas-side boundary condition in raising the mean effective gas-temperature is clearly shown. The present matching method can thus be used in the further investigation of the effect of insulation on the performance of the whole compound engine system.

Date of Award	1982
Original language	English
Awarding Institution	University of Bath