Experimental and numerical investigation of transient starting of pre-evacuated exhaust diffuser in high altitude ground test

Abstract

Listen

Translate article

This paper deals with numerical and experimental investigation of exhaust diffuser transient starting in a high-altitude ground test of a thrust optimized parabolic (TOP) nozzle. The experimental investigation has focused on cold gas flow inside TOP nozzle and second throat exhaust diffuser (STED), performed to examine flow unsteadiness at the starting phase. In order to identify physical phenomena during STED starting, unsteady numerical simulations have also been carried out. Two major parameters including nozzle pressure slope and pre-evacuated regions along the diffuser have been examined. The results indicate elimination of the cap shock separation or restricted shock separation (RSS) pattern during the transient starting at high nozzle pressure slope with pre-evacuation. Also, considerable reduction in the starting time is achieved, so that it causes more desirable instantaneous starting of the diffuser in some conditions. However, at average and lower nozzle pressure slopes, RSS pattern still develops inside the nozzle. Investigation of the effect of pre-evacuation spaces along the diffuser on the course of evacuation development indicates that the larger the pre-evacuation space along the diffuser, the more desirable is the STED operation quality, whereby the starting time decreases considerably. It is shown that with a given nozzle pressure slop, enlarging the pre-evacuation area from the diffuser inlet to the middle of the second throat results in a 25% reduction in diffuser starting time, and the further increase of the pre-evacuation area to the end section of the diffuser leads to instantaneous starting of the diffuser. Furthermore, we found that initial pre-evacuation pressure value has less effect on the diffuser starting process, so that for a given nozzle pressure slop and a pre-evacuation area by reducing this parameter five times, the starting time is reduced by only 3.75%.