Friction factor and heat transfer coefficient behavior are investigated experimentally under mixed convection conditions in aiding and opposing transition and turbulent flow of water (4000 < Re < 9000 and Bo < 1.3). With increasing buoyancy influence, the friction factor increases by as much as 25 percent in aiding flow, while it decreases by as much as 25 percent in opposing flow (GrΔT < 7·106). The effects of temperature-dependent viscosity variations are also included in the analysis (0.5 < μw/μb < 1.0). When they are taken into account, the increase in the friction factor due to buoyancy forces alone in upward flow becomes larger. The friction factor behavior is compared with previous studies in the literature. Our experimental data agree well with some of the previous experiments described in the literature. The heat transfer coefficient was also measured under the same experimental conditions; the heat transfer coefficient monotonically increases in opposing flow by as much as 40 percent, and first decreases by 50 percent and then recovers in aiding flow with increasing buoyancy influence.

A two-dimensional numerical model of the CANDU Moderator Test Facility (MTF) is presented. The results show fluctuations in velocity and temperature. The thermal solution of the MTF model indicates that the buoyancy forces dominate in the inner core of the tank, whereas, the inlet jet induced inertial forces dominate the outer edges of the tank. The interaction between these flows forms a complex and unstable flow structure within the tank. The largest flow fluctuations occur outside the tube bank where the inlet jets continue to flow, and around the top of the tank where the two inlet jets impinge on each other. The high velocity gradients between the inlet jets and the surrounding slower moving moderator fluid generate small vortices with low fluctuation amplitude but high frequencies.

A simple method has been developed to model boiling heat transfer from a heat exchanger to pools using the experimental data available in the literature without modeling the flow dynamics of the pool. In this approach the heat flux outside vertical tubes is expressed as a function of outside wall temperature of the tubes and saturation temperature of the pool at or near atmospheric pressure.

RELAPS MOD3.1.1 is being used to simulate Loss of Coolant Accidents (LOCA) for the Simplified Boiling Water Reactor (SBWR) being proposed by General Electric (GE). One of the major components associated with the SBWR is the Passive Containment Cooling System (PCCS) which provides the long-term heat sink to reject decay heat [l]. The results of a set of RELAP5 calculations at these conditions were compared with the GIRAFFE data. The effects of PCCS cell nodings on the heat transfer process were also studied. The UCB correlation. as implemented in RELAPS, predicted the heat transfer to f 5 % of the data with a three-node model. The three-node model has a large cell in the entrance region which smeared out the entrance effects on the heat transfer, which tend to overpredict the condensation. Hence, the UCB correlation predicts condensation heat transfer in the presence of noncondensable gases with only a coarse mesh. The cell length term in the condensation heat transfer correlation implemented in the code must be removed to allow for accurate calculations with smaller cell sizes.

Safety analysis computer codes are designed to simulate phenomena relevant to the assessment of normal and transient behaviour in nuclear power plants. In order to do so, models of relevant phenomena are developed and a set of such models constitutes a computer code. In accident or transient analysis the values of certain output parameters (margin parameters) are used to characterize the severity of the event. The accuracy of the computer code in calculating these margin parameters is usually obtained through validation and variation in the margin parameter is estimated through the propagation of variation in the code input. A method for estimating code uncertainty respect to a specific output parameter has been developed. The methodology has the following basic elements: (1) specification and ranking of phenomena that govern the behaviour of the output parameter for which an uncertainty range is required; (2) identification of models within the code that represent the relevant phenomena; (3) determination of the governing parameters for the phenomenological models and Identification of uncertainty ranges for the governing model parameters from validation or scientific basis, if available; (4) decomposition of the governing model parameters into related parameters; (5) identification of uncertainty ranges for the modelling parameters for use in Best Estimate Analysis; (6) design and execution of a case matrix; and (7) estimation of the code uncertainty through quantification of the variability in output parameters arising from uncertainty in modelling parameters. This methodology has been employed using simulations of Large Break Loss of Coolant Accident (LOCA) tests in the RD-14M test facility to calculate the uncertainty in the TUF thermal hydraulics code calculation of the coolant void fraction. The uncertainty has been estimated with and without plant parameters (parameters specific to the RD-14M test loop). The TUF coolant void fraction uncertainty without plant parameters was determined to be 0.08 while the uncertainty with plant parameters included was determined to be 0.11. The uncertainty value without plant parameters included is comparable to the uncertainty in the measurements (0.09). The uncertainty value with plant parameters included is larger than the variation in the bias (0.10) of the TUF calculation of void fraction. From these findings, it can be concluded that the estimated accuracy of the TUF code calculation of void fraction is consistent with the available experimental data.