Investigating the Unsteady Flow Physics within a Mach 1.8 Isolator

Engineering tools with proven capability to diagnose scramjet isolator design challenges, especially tools that will prevent unstarts, are lacking. To this end, a numerical model was developed to investigate the unsteady flow physics within a 2D, Mach 1.8 isolator with a length to height ratio of 8.40. The Integro-Differential Scheme (IDS) was developed in (Ref. [1]) was used in this analysis. The aerodynamic conditions used in the design of the numerical model was extracted from the experimental data presented in (Ref. [2]). The flow physics within the isolator numerical model was studied under varying back pressure conditions. The exit pressure value, PExit, computed under the ‘prescribed design conditions’ using supersonic outflow conditions was considered the ‘baseline back-pressure’, and it was defined by the symbol, PBL. Using this definition, the flow physics within the isolator was studied for a range of back pressure, PBP, defined by PBP = aPBL, where is a non-dimensional design parameter. In this study a varied from 1.0 to 2.5, at intervals of 0.1. In each case, for a fixed a, the full unsteady Navier-Stokes equations were simulated for over 20 cycles (20tau), where each cycle represented the average time, tau, a given fluid element takes to traverse the full length of the isolator. It is noteworthy to also mention, backpressure studies were conducted through the use of ‘smooth’ and ‘discrete’ pressure jumps. The engineering analysis conducted herein demonstrated results that are in excellent agreement with the available experimental data. Moreover, the results obtained from this analysis demonstrated that the 2D IDS computational fluid dynamic tool accurately predicted the fluid physics within the isolator. It was observed that under design conditions, the isolator flow field consisted of an oblique shock train, which was strongest closest to the entrance of the isolator. Also, it was observed during each ‘discrete’ change in back pressure value, a wave, comprising of a coupled pair of oblique shocks and a normal shock, resembling the ‘lambda shock pattern’ emerges from the exit of the isolator. During each test, this ‘lambda shock’ travels to the front of the isolator, interacting with and dominating each set of reflected waves along its path. In each case, the lambda shock interacts with the front-most and strongest pair of oblique shocks, rocking back and forth before the entire isolator flow field settles down into a new configuration. This process intensifies as the back pressure increases in strength, and the oblique shock train transformed into a form that closely mimics a normal shock train, with the strongest ‘lambda shock’ at the head of the isolator. In general, it appears as if the isolator flow patterned itself as a flexible spring within the constant area duct, constantly modifying its ‘net shock strength’ to accommodate the rising back pressures and while allowing the leading lambda shock small increments towards the entrance. The results showed that at a PBP with a = 2.1, the leading lambda shock moves rapidly towards the entrance, and with a PBP with a = 2.2 the isolator ‘unstarts’.