Numerical Evaluation of the Performance of Supersonic Exhaust Diffuser with and Without Ejected Mass

A numerical investigation has been carried out on straight cylindrical supersonic exhaust diffusers (SED) of a high-altitude testing (HAT) to study the nature of vacuum generation with mass ejection from the test cell. Simulations were carried out for SED operating in self-evacuation mode and, SED operating in assisted-evacuation mode (by means of mass ejection). In HAT facilities, the nozzle back-pressure is artificially maintained at a suitably lower level in order to determine the vacuum thrust, maximum heat loads, transient conditions such as engine startup, shut down and re-ignition in vacuum conditions. HAT maintains the required vacuum by arresting the backflow from the atmosphere into the altitude simulation chamber. The steady SED operation is characterized by two distinct operating modes: diffuser unstarting and diffuser starting mode. In the started condition, the vacuum chamber attains its highest vacuum state. Hence, in the case of HAT the most important considerations are the minimum starting and operating pressures. The basic mechanism of SED is the flow of a compressible fluid in a confined duct under the influence of a strong adverse pressure gradient. Wall flow separation under adverse pressure gradient and shocks often results in large recirculation bubbles appearing in SED. Often these recirculation bubbles move towards the low-pressure test cell, preventing the further entrainment of mass from test cell, thereby adversely affecting its vacuum performance. Large improvements in SED performance are possible by controlling or arresting the backflow. The academic community has extensively explored several backflow controlling devices like orifice plate at the inlet of SED, backflow arrester device in vacuum chamber, nozzle lip thickness etc. It is reported that the vacuum ejector start-up is associated with a gradual initial vacuum generation stage, which is followed by a rapid vacuum generation. The slow phase of initial vacuum generation observed in SED can be accelerated by mass ejection. Previous experimental studies with mass-ejection from the vacuum chamber have reported that when mass was ejected from the test-cell, lower pressure was achieved. However, little research has been conducted to show the effect of mass ejection on the starting transients and altitude simulation performance of an ejector.