Control strategies for stabilizing transpiration cooling with phase change under high heat flux conditions

Phase-change transpiration cooling is a highly effective thermal protection technique for hypervelocity vehicles but suffers from inherent instabilities such as cooling hysteresis and temperature oscillations. This study presents an experimental investigation of closed-loop automatic control for phase-change transpiration cooling using both pump-driven and gas-pressurized coolant delivery systems. The transpiration cooling system consists of a sintered bronze porous plate as the transpiration specimen, a stainless steel fixture (including a water reservoir and two flow straighteners beneath the porous plate), a solenoid valve, temperature and flow rate measurement and control systems, a deionized water tank or water bladder, and either a pump or a gas-pressurized system as the coolant driving source. A PID control algorithm is simultaneously implemented to regulate the temperature of the porous surface and the coolant flow rate. The incorporation of a water-retention material significantly enhances system stability by improving the self-pumping capability of the porous medium. A gas-pressurized coolant delivery system, designed to meet the structural and weight constraints of practical spacecraft applications, is developed and characterized under high heat flux conditions. Experimental results show that the water-retention material substantially improves system dynamic behavior, enabling stable temperature control with fluctuations below ±1%. Flow rate constraints significantly influence control performance: a wider range (e.g., 1–10 g/min) enhances response speed but introduces larger pressure oscillations, while initiating control near the steady-state operating point promotes stabilization. For heat flux step changes (2 MW/m 2 → 2.5 MW/m 2 ), a broader flow rate limit reduces the response time by 97.4 s and the peak temperature overshoot by 9.4 °C. Prior knowledge of the impending step combined with a 10 s lead time in setpoint reduction further reduces the response time by 45.6 s and the overshoot by 16.8 °C.