Development and Modelling of Expansion Tubes

The accurate simulation of flow conditions encountered by aircraft is imperative to the development of aeronautics and astronautics. Test flights are expensive and time consuming, thus, ground based test facilities provide the majority of data in new flight regimes. A mixture of analytical studies, Computational Fluid Dynamics (CFD), experimental results from ground based test facilities and actual atmospheric flights are required to obtain important design parameters. Coupled with this need for aerodynamic information, is a constant desire to simulate higher flow speeds. Expansion tubes provide an option for testing aeroshells such as the new Crew Exploration Vehicle (CEV) or the high enthalpy end of scramjet flight regime. A unique opportunity already exists in which the fabricated hardware for the RHYFL shock tunnel could be utilized to create the world’s largest expansion tube, RHYFL-X. This thesis continues a project which attempted to investigate if RHYFL-X would be a beneficial experimental tool. This thesis began by installing and commissioning a new single-stage, free-piston driver for the X2 expansion tube located at the University of Queensland. The new configuration provided a smaller scale version of the RHYFL-X expansion tube on which testing could be performed. It also provided a number of advantages which would lead to increased performance in the X2 facility. Once installation was complete, a series of experiments was conducted to ensure that the new driver configuration was operating correctly and to determine the capabilities of it. Four conditions were examined. Two air conditions which produced secondary shock speeds of 8.14 km/s and 8.61 km/s, a condition to simulate the entry into the Titan atmosphere at 5.51 km/s and another condition to simulate entry into the Jovian atmosphere at 10.2 km/s. Numerical one-dimensional models were examined along with hybrid simulations, which used the one-dimensional model up to rupture of the secondary diaphragm and then coupled this to an axisymmetric simulation of the acceleration tube. The results from these numerical simulation models compared well to the experiments and showed that the new driver provided a significant increase in the stagnation pressures that could be generated in the expansion tube over the old two-stage, free-piston driver. There have been claims in previous studies that an expansion tube nozzle would not only increase the core flow diameter but also the steady time available for testing. However, this claim has never been physically demonstrated. A full-capture, contoured, shockfree nozzle was designed for the new X2 configuration. The design code for the nozzle incorporated a flow solver for the Parabolized Navier-Stokes (PNS) equations coupled to a Nelder-Mead optimization algorithm. A 1.4m nozzle was installed at the end of the acceleration tube and a series of experiments conducted. CFD simulations similar to the hybrid method mentioned above were conducted with the nozzle attached. The experiments again matched well with the numerical models. The results indicated that the addition of the nozzle produced a larger core flow diameter and longer test time as had been claimed. However, the improvements were not as significant as predicted by the axisymmetric simulations. Extremely large boundary layers were formed but sufficient core flow was still generated to enable the testing of models. Both the addition of a single-stage, free-piston driver and a contoured nozzle has significantly increased the capabilities of the X2 expansion tube. The numerical models that are used in this thesis provide an excellent basis for the prediction of the flow that can be generated in the X-series of expansion tubes located at the University of Queensland. Following the success of this project, the larger X3 expansion tube located at the University of Queensland is going to be refurbished with a single-stage, free-piston driver and a nozzle.