The current decade has seen a major effort by European countries to achieve competitive status in the field of com-munication satellite systems. Effort and investment has been channelled by governments either nationally (the Italian Sirio Programme), bilaterally (the Franco-German Symphonie Programme) as well as through the multinational ESA, the European Space Agency (the OTS/ECS and Marecs Programmes). Europe is now at the crossroads of moving from the technological and preoperational phase, that has resulted from such programmes, to the implementation of operational systems responding to its own requirements. In addition, a technological base has been created in European industry capable of enabling it to compete for the worldwide market for regional and domestic systems that is now developing.

Nearly 15 methods have been proposed to obtain the shape of an axisymmetric wind tunnel contraction, but only a few have been studied experimentally. Bibliographic details of these methods can be found in Refs. 1, 2 and 3. Of these, the method described by Thwaites has the advantage that the contour and the velocity distributions are given directly by simple series expression. This method also has the criticism that the solution is periodic and does not satisfy near uniform stream condition obtained at inlet and outlet. But such a criticism applies to all methods which assume the length of contraction to be finite (e.g. methods due to Batchelor and Shaw, Chmielewski. Further, an experimental investigation by Klein et al(3) showed that Thwaites’ method with first four terms in solution gave a better performance than the contour given by Batchelor and Shaw.

It was proposed in Ref. 1 that cryogenic temperatures can be achieved by the adiabatic isentropic expansion of a gas. The scheme is shown in Fig. 1. High pressure gas contained in A expands by means of valve 1 and cools. Valve 2 is then opened along with nozzle valve to push cold test gas through the working section.

In the ideal case the process is isentropic. To achieve reservoir conditions of T=120K and P = 4·4 bars, the initial pressure needed would be 110 bars. However, because of heat transfer from the walls of the tube, the process is not ideal and the actual initial pressure needed is higher. The main parameters which affect the heat transfer are the internal surface area of the tube, the rate of expansion governed by the size of valve 1 and the initial pressure of the gas.