The flow of saturated and near-saturated water through safety relief valves

This investigation centered on the flow of flashing water in nozzles and nozzle type safety relief valves widely used in steam generating plants and in the chemical process industry. These valves often protect vessels which contain liquids that will flash when vented to the atmosphere. Their operation in two-phase flow is not well understood. The valves range in throat area size from 0.110 to 26 square inches. The exit areas are not proportional to the throat areas, the ratio varying from 17 for the smallest valve to 3 for the largest. This study undertook to determine: if the methods for predicting two-phase flows through nozzles could be applied to valves; if the exit-to-nozzle-area ratio changed the mass flow rate and/or caused valve instability; and if the ASME and API codes provided proper guidance regarding valve sizing.;A blowdown type test apparatus was built in which both valves and nozzles were tested with water and steam at temperatures up to 350 degrees F and at pressures up to 120 psia. Bushings were installed in the valve exit to simulate the area ratio of the largest valves.;As a result of this study a new semi-empirical computational model was developed which is based on few assumptions, is physically realistic, and is computationally easy to use. It was demonstrated that under the conditions tested, the two-phase flow rates for nozzles and nozzle type relief valves were indistinguishable. Exit-to-nozzle area ratio was found to be an important valve parameter. Under saturated conditions, at an area ratio corresponding to the largest valves, the conventional valves showed a 40 percent decrease in mass flux as compared to smaller valves of similar design, and a low frequency instability (0.8 to 3 Hz) indicating choking at the exit. This decrease in mass flux and the occurrence of instability, resulting in periodic valve closure, persisted at temperatures down to 50 degrees of subcooling.;The API Recommended Practice, which is based on a fictitious two fluid model, gives good predictions in two-phase flow, except when there is choking due to low exit-to-nozzle-area ratio. The ASME Pressure Vessel Code does not recognize the importance of subcooled flashing and consequently following its recommendations literally will result in mass flow rate predictions which are 100% higher than the actual flow rates. At saturation, however, the ASME Code predictions are based on the HEM procedure which for nozzle type valves leads to mass flow predictions which are approximately 50% of the actual flow rates. The ASME code, like the API, does not recognize the possibility of exit choking.