A Model for Computing Two-Phase Pressure Drop in Vertical U-Tube Steam Generators and its Application to Thermosyphoning

Abstract

This paper presents a steady state model, dubbed THERMOSYPHON, for a quantitative analysis of two-phase thermosyphoning in the heat transport system of a facility with multiple horizontal channels and inverted U-tube steam generators arranged in a geometric configuration similar to that in CANDU. Thermosyphoning is a mode of heat removal where the heat from the channels is transported to the steam generators by either single- phase or two-phase natural circulation flow over the top of the U- tubes. The model is developed to primarily isolate and understand the important physical mechanisms underlying the phenomena in thermosyphoning. Another objective of the model is to provide a fast-running tool for a parametric study of thermosyphoning. Thus, THERMOSYPHON complements the larger thermalhydraulic codes. THERMOSYPHON computes the steady state circuit flow and pressure, the temperature and coolant phase distribution in the U- tubes, and pressure difference between the inlet and outlet of the steam generators and between the inlet and outlet headers. These parameters are computed as a function of the loop integrated void fraction, steam generator secondary side pressure, and heat losses. THERMOSYPHON also computes the loop conditions for the onset of circuit flow reduction, any thermosyphoning breakdown, sustained reversal in the direction of channel flows, and any channel heatup. The model uses only one (average) U-tube which is argued to be adequate. It is proposed with some justification that above a certain loop void and during a loop depressurization resulting from either loop cooldown or loss of inventory, coolant flashing is more likely at the top and in the cold leg than in the hot leg of the steam generator U-tube. This flashing would occur even for relatively high values of primary-to-secondary side fluid temperature difference. THERMOSYPHON predicts the coolant phase distribution in the U-tube and the differential pressure across the steam generator resulting from this cold-leg flashing. The cold-leg flashing is shown to cause the significant increase in the differential pressure across the steam generators, the channel flow reversal, thermosyphoning breakdown, and channel heatup. The predicted values of the above loop parameters as well as the conditions for the onset of channel flow reversal and heatup agree reasonably well with those measured in one of the high pressure two-phase thermosyphoning experiments conducted in the multiple channel RD-14M test facility. THERMOSYPHON predicts that, in a loop such as CANDU with small heat losses, effective thermosyphoning would persist to higher values of loop void than those observed in the high pressure two-phase thermosyphoning RD-14M experiments.