<p>  </p> <p>High altitude test facilities are needed for ground testing of upper stage rocket engines or small satellite thrusters with high expansion ratio nozzles to ensure full-flowing nozzle conditions. Rocket exhaust diffusers and ejector systems are essential components of these facilities and are frequently used to set desired simulated altitude/low pressure conditions and pump out rocket exhaust products. </p> <p>This dissertation combined experimental and numerical efforts on diffuser-ejector systems. The experimental efforts included the development of a Second Throat Exhaust Diffuser (STED) to aid with the qualification of space thrusters in the Purdue Altitude Chamber Facility. While performing these experiments, we characterized the single and two-stage ejector systems operating in conjunction with the diffuser to obtain and maintain specific simulated altitudes. </p> <p>The concurrent numerical effort focused on validating a Computational Fluid Dynamics (CFD) approach based on Reynolds-averaged Navier–Stokes equations flow simulations. After validating the ejector CFD, we used it to derive a corrective coefficient of a lumped parameter ejector model (LPM) developed previously for the ejectors used in the Purdue Altitude Facility. We created variable coefficient maps for the stages of the two-stage ejector system using the same LPM and the test data from one of our experiments. </p> <p>We designed, manufactured, and then validated a STED for altitude testing of a ~50 lbf hypergolic hybrid motor as a part of a NASA JPL project. The designed STED enabled the operation of the hybrid motor for the full duration of the test firing (about 2 seconds) at a simulated altitude of 102,000 feet, slightly above the targeted altitude of 100,000 feet. We also validated our diffuser CFD approach by creating a simulation using the measured diffuser back pressure and the average motor chamber pressure. </p> <p>We then devised an experiment to investigate several diffuser–ejector system configurations using cold gas thrusters with conical and bell nozzles. The main aim of that experiment was to explore the effects of different thruster nozzle geometries, diffuser geometries, and thruster/ejector operational parameters on the performance of a diffuser–ejector system. For all the configurations tested, we reported on the minimum starting and operating pressure ratios and corresponding correction factors on the normal shock method. The large hysteresis regions obtained mostly with bell nozzles having a high initial expansion angle provided an opportunity to economize the facility resources. In some cases which were later found to violate STED second throat contraction limits, we experienced a choking flow at the second throat. Then, we studied the second throat contraction limits in detail using CFD in addition to the experimental data and explored minimum diffuser second throats enabling diffuser starting and improving aerodynamic efficiency. </p> <p>Finally, we machined a larger scale cold gas thruster with different nozzle geometries (having throat diameters in the range of 0.367 – 0.52 inches) from acrylic rods to study possible flow separation and gas condensation events that could occur during tests in the altitude chamber. The main difference here with the previous experiment was that the diffuser (JPL STED) was fixed, and the two-stage ejector system was used to create the necessary back pressure. With the experiments performed at varying axial gaps between the nozzle exit and diffuser inlet, we were able to investigate the effect of that on the diffuser performance. The experimental data collected in this work and the complementary numerical efforts served to generate the operating envelope of the Purdue Altitude Chamber Facility.  </p>