No abstract

The paper describes a theoretical and experimental investigation into unsteady flows in simple branch systems. In order to simplify the analysis one-dimensional unsteady flow theory is used to analyse the wave action in the pipes. The method of characteristics is used for this purpose, and both homentropic and non-homentropic flows are investigated.The boundary conditions at the branch are represented by quasi-steady flow theories. Two such theories are examined, the first called the constant pressure theory assumed no pressure change across the branch at the junction. A full graphical treatment is given elsewhere (4). In the paper the assumptions and the formulation of the equations for an electronic computer are briefly discussed. The second method assumes a pressure drop across the branch at the junction. An approximate theoretical method of estimating the pressure drop with certain types of flow is given. Using experimentally determined steady flow loss coefficients at the junction the expressions used in the second method are formulated for direct computation using a digital computer.The results are given of an extensive experimental investigation to test the validity and limits of the theories described above. Steady flow experiments to evaluate the loss coefficients at the junction over the range of pressure and flow directions covered by the unsteady tests are described. The unsteady flow tests on a special multi-cylinder pulse generator are discussed in detail. The results show that under certain conditions the calculated pressure diagrams using constant pressure theory do not agree with the experimental diagrams, both the pulse shape and amplitude being incorrect. Using the pressure drop theory together with experimentally determined steady flow loss coefficients there is considerable improvement in the theoretical diagrams.Agreement is obtained over a wide range of flows.

Vinyl acetate (a polar compound) has become a key ingredient in many industrialand consumer products. However, very little is known about its thermodynamic properties. This article presents a simple modijication of Wu and Stiel's (1 985) method (original/) developed for predicting thermodynamic properties of polar compounds up to P, = 10) for extending its range of applicability to very high pressures of the order of P, = 80. The predictions show good agreement with the experimental data for specific volume in the case of vinyl acetate, and specific volume, enthalpy and entropy for mixtures of ethylene and vinyl acetate for compositions up to 50 wt. % of vinyl acetate. Further validation for other polar systems would determine if this method is applicable to polar compounds in general.

No abstract

A numerical method has been developed for the solution of one-dimensional non-steady flow problems including the effects of friction, gradual area change, heat transfer between the gas and the pipe wall, longitudinal variations, and discontinuities in gas composition and entropy. The fluid considered obeys the perfect gas equation of state, but the specific heats may vary with temperature. The method is not intended for use when shocks are present, but will give an approximate solution if shocks occur. The accuracy of the new method has been checked against existing methods for more simple cases, and although the new method has been found to be slightly superior it is more complicated, much slower, and the boundary conditions are more difficult to develop. For this reason, it is suggested that the new method be used to check on the adequacy of the existing simpler methods for each new application. The methods have been compared for the case of a typical diesel engine exhaust system and it has been found that the earlier methods are adequate for all practical purposes.