This paper provides single and two phase rod bundle data to support verification of heat transfer models being used in steaming rate and crud model predictions for rod bundles. The effort to summarize this work was supported by the EPRI Robust Fuel Program and is defined in more detail in EPRI report 1003383. Subcooled boiling tests were performed by Combustion Engineering (CE) in early 1980s to provide insight on heavy crud deposits and fuel failures observed on peripheral rods for bundles in Maine Yankee cycle 4. Two 5×5 tests were performed at the Columbia University Heat Transfer Research Facility simulating the peripheral region of adjacent CE 14×14 fuel bundles for two different perimeter strip geometries. The test conditions were at typical reactor pressure, temperature, and heat flux. The rods were 7’ in heated length and were electrically heated with a uniform axial power shape. There were no mixing vanes on the spacer grids. Thermocouples were placed on the hot rod in the center of the test section and on an adjacent rod at 4 different axial levels. Thermocouples were also located in the center of the subchannels at the end of the test section. Boiling curves were generated over a range of test conditions (system pressure, inlet temperature, and flow rate) by plotting rod surface temperature versus heat flux. The boiling curves covered single phase, subcooled boiling, bulk boiling and DNB conditions. The data from the boiling curves were reduced and evaluated with the VIPRE thermal hydraulic code. Clad temperature predictions were made with VIPRE based on available heat transfer correlations for comparison to clad temperature measurements. These heat transfer correlations include the Dittus Boelter correlation for single phase flow, the Jens Lottes, Thom and Chen correlations for two phase flow conditions (subcooled boiling). The VIPRE predictions of the hot rod average surface temperature, based on the Dittus-Boelter correlation with a grid enhancement factor for single-phase forced convection and the Thom correlation for nucleate boiling, gave the best agreement with the rod bundle test data among all the available modeling options. It was concluded that current heat transfer models used in TH codes, are adequate for average steaming rate calculations supporting Axial Offset Anomalies (AOA) evaluations, as long as the appropriate grid enhancement factor is utilized for the spacer grids in the analysis. However, further testing and modeling may be needed to simulate local grid effects and hot spots downstream of spacer grids.

VERA-CS (Virtual Environment for Reactor Applications, Core Simulator) is a coupled neutron transport and thermal-hydraulics subchannel code under development by the Consortium for Advanced Simulation of Light Water Reactors (CASL). VERA-CS was applied to simulate core behavior of a typical Westinghouse-designed 4-loop pressurized water reactor (PWR) with 17x17 fuel assemblies in response to two main steam line break (MSLB) accident scenarios initiated at hot zero power (HZP) at the end of the first fuel cycle with the most reactive rod cluster control assembly stuck out of the core. The reactor core boundary conditions at the most DNB limiting time step were determined by a system analysis code. The core inlet flow and temperature distributions were obtained from computational fluid dynamics (CFD) simulations. The two MSLB scenarios consisted of the high and low flow situations, where reactor coolant pumps either continue to operate with offsite power or do not continue to operate since offsite power is unavailable. The best estimate plus uncertainty (BEPU) analysis method was applied using Wilks' nonparametric statistical approach. In this demonstration of BEPU application, 59 full core simulations were performed for each accident scenario to provide the minimum departure from nucleate boiling ratio (MDNBR) at the 95/95 (95% probability with 95% confidence level) tolerance limit. A parametric goodness-of-fit approach was also applied to the results to obtain the MDNBR value at the 95/95 tolerance limit. Initial sensitivity analysis was performed with the 59 cases per accident scenario by use of Pearson correlation coefficients. The results show that this typical PWR core retains design margin with respect to the MDNBR safety limit for both of the MSLB accident scenarios. The scenario with available offsite power was more restrictive in terms of MDNBR than the scenario without offsite power.

The Virtual Environment for Reactor Applications core simulator (VERA-CS) being developed by the Consortium for Advanced Simulation of Light Water Reactors (CASL) includes coupled neutronics, thermal-hydraulics (T/H), and fuel temperature components with an isotopic depletion capability. The neutronics capability is based on the Michigan Parallel Characteristics Transport Code (MPACT), a three-dimensional whole-core transport code. The T/H and fuel temperature models are provided by the COBRA-TF (CTF) subchannel code. As part of the CASL development program, the VERA-CS (MPACT/CTF) code system was applied to model and simulate reactor core response with respect to the departure from nucleate boiling (DNB) ratio at the most limiting point of a postulated pressurized water reactor main steam line break event initiated at the hot zero power, either with offsite power available and the reactor coolant pumps in operation (high-flow case) or without offsite power, where the reactor core is cooled through natural circulation (low-flow case). The VERA-CS simulation was based on core boundary conditions from the RETRAN-02 system transient calculations and STAR-CCM+ computational fluid dynamics (CFD) core inlet distribution calculations. The evaluation indicated that the VERA-CS code system is capable of modeling and simulating quasi-steady-state reactor core response under the main steam line break accident condition, the results are insensitive to uncertainties in the inlet flow distributions from the CFD simulations, and the high-flow case is more DNB limiting than the low-flow case.

COBRA-TF (Coolant Boiling in Rod Arrays – Two Fluid) or CTF is a transient subchannel code, selected to be the reactor core thermal hydraulic (T/H) simulation tool in the multi-physics code development project of the Consortium for Advanced Simulation of Light Water Reactor (CASL) sponsored by the US Department of Energy. CTF is currently being evaluated and further improved by CASL as part of its multi-physics software package to help the nuclear industry address operational and safety challenge problems, such as Departure from Nucleate Boiling (DNB) and Reactivity Initiated Accidents (RIA). In this paper, CTF’s capability for transient fuel thermal analysis, including DNB prediction is evaluated by modeling and simulating power burst experiments with high burnup PWR fuel rods, conducted at the Nuclear Safety Research Reactor (NSRR) in Japan. The experiments were a series of tests performed using pulse irradiation capability of the reactor to evaluate fuel rod failure with respect to fuel enthalpy, coolant conditions, and fuel design during RIAs such as control rod ejection. Specific to this study, the experiments using the Takahama-3 reactor fuel segments have been modeled and simulated to evaluate CTF’s prediction capability for DNB onset, fuel rod thermal response, and heat transfer from single-phase to post-CHF during fast RIA transients.

No abstract