Experimental investigation on stable/unstable flow behaviors of parallel boiling channels under forced vertical vibrations

Abstract

• Stable/unstable boiling flows in parallel channels were tested under vibrations. • A non-dimensional void difference was introduced to identify system stability. • Stable flow behaviors can be clearly affected by forced vertical vibrations. • Unstable flow oscillation frequency can be validated with 1-D drift flux model. • An empirical correlation for predicting oscillation frequency was proposed. Experimental tests were carried out to investigate the dynamic behaviors of parallel boiling channels under stable and unstable flow conditions with/without forced vertical vibrations. The parallel channels with a length of 1.91 m and diameter of 11.9 mm were attached to a vertical vibration platform. The inlet flow was controlled with average velocity of 0.142–0.277 m/s and subcooling of 6.0–12.6 °C, and the parallel channels were loaded with heat flux of 12.5–36.9 kW/m 2 . Overall 35 flow conditions were tested along with three static and vibration subset conditions ( f = 0, 1.06 and 1.62 Hz), which covered the stable and unstable flow conditions in the parallel channels, and the dynamic variations of void fraction, pressure difference and temperature were recorded and analyzed. A non-dimensional void difference was proposed to identify the system stability of parallel channels, and fast Fourier transform (FFT) was utilized to extract the dominant frequencies of flow behaviors. While system is stable under vibrations, the dominant frequencies of flow properties were identical to the vibration frequencies; whereas if the system is unstable, the extracted frequencies were dominated by unstable flow oscillations, which are much lower than the vibration frequencies. In addition, the unstable flow oscillation frequencies can increase with increasing inlet velocity and heat flux, and an empirical correlation was proposed for estimating the oscillation frequencies, which can predict the present database with an averaged accuracy of 8.85%.