This paper describes the modeling and control of a mixing chamber used in rocket engine testing at the NASA Stennis Space Center. The mixer must combine high pressure liquid hydrogen (LH2) and gaseous hydrogen (GH2) to produce an output flow that meets certain thermodynamic properties before it is fed into a test article. More precisely, this paper considers that the quantities to be tracked and/or regulated are mixer internal pressure, exit mass flow, and exit temperature. The available control inputs are given by three valve positions, namely those of the GH2, LH2 and exit valves. The mixer is modeled by a system of two nonlinear ordinary differential equations having density and internal energy as states. The model must be simple enough to lend itself to subsequent feedback controller design, yet its accuracy must be tested against real data. For this reason, the model includes function calls to thermodynamic property data. Some structural properties of the resulting model that pertain to controller design, such as controllability and uniqueness of the equilibrium point are shown to hold. As a first control approach, a small-signal (linear) model is developed based on the nonlinear model and simulated as well. Pulse disturbances are introduced to the valve positions to examine the effect of operator manual operation. Valve control strategies that simulate an operatorin- the-loop scenario are then explored demonstrating the need for automatic feedback control. Classical optimal single-output and multi-output Proportional/ Integral controllers are designed based on the linear model and applied to the nonlinear model with excellent results to track simultaneous, constant setpoint changes in desired exit flow, exit temperature, and mixer pressure, as well as to reject unmeasurable but bounded additive step perturbations in the valve positions. A feedback linearization controller is designed and used to achieve tracking and regulation of the outputs over an extended range of the variables of interest. It is shown that the system is minimum-phase provided certain conditions on the parameters are satisfied. The conditions are shown to have physical interpretation. The control strategies are tested by implementing the nonlinear differential equations in the SIMULINK environment including a table look-up feature of the fluid thermodynamic properties. [12, 4, 5, 2]. © 2003 by the American Institute of Aeronautics and Astronautics, Inc.