K.B. Cady

A Procedure for Evaluating Modified Pulsed-Neutron- Source Experiments in Subcritical Nuclear Reactors

AbstractA theoretical interpretation of the modified pulsed-neutron-source experiments of Sjostrand, Gozani, and Garelis and Russell is given using exact steady-state Boltzmann equations. The interpretation is based on a phenomenological description of the experiments and is patterned after work done on the Garelis-Russell method by Corngold. The basic approximation made is that the fundamental prompt-mode decay constant is much larger than any delayed-neutron precursor decay constant. The theoretical interpretation allows the reactivities measured by the above three modified pulsed-source techniques to be easily calculated and compared to more conventional definitions of reactivity. The calculations can be performed by any standard source-iteration code that has been modified to solve the inhomogeneous problem. Experiments were performed on the Cornell University Critical Assembly and interpreted with the aid of the above theory. Calculations and experiments agree to within 20%. Sjostrands method is fou...

Performance and calibration of a neutron image intensifier tube based real-time radiography system

Image calibration is central to extending the capabilities of neutron radiography beyond mere visualization. However, the effects of scattered neutrons and variations in background image intensity adversely affect quantitative radiography. We describe the calibration of a real-time neutron radiography system that limits these effects and which is applicable to systems with variable digitizer gain and offset. A neutron image intensifier tube coupled to a vidicon camera with a capture rate of 30 frames/s was used. The system could account for 10 ml of water entering the field of view to within 2% and could measure the variation in thickness of a graphite wedge to within 2.3%. The spatial resolution was 450 /spl mu/m for a field of view of 410 cm/sup 2/. The image persistence half life was /spl sim/0.3 s and the system was functional for quantitative radiography with neutron fluxes above /spl sim/5*10/sup 5/n/cm/sup 2//s.

Dynamical models and numerical simulation of system-wide transients in loop-type LMFBRs

Abstract Dynamical models and numerical methods for a digital simulation of protected transients in loop-type LMFBRs resulting in EPRI-CURL code are presented. The model is capable of simulating operational transients, anticipated incidents, and postulated accidents which do not lead to sodium boiling. The dynamical models include: point reactor kinetics, primary, intermediate, and tertiary system heat transfer and coolant flow dynamics governed by forced and natural convection effects; and plant protection and control systems. A numerical method is incorporated which calculated the characteristic times of the 489 state variables modeling the entire system, and compares them with a variable preset integration timestep. A Runge-Kutta algorithm is applied to those state variables with moderate and slow response, and a quasistatic approximation is applied to those with rapid response; i.e., the ‘stiff’ equations. This assures numerical stability and is shown to greatly reduce the computation time requirements without much sacrifice in accuracy. The steady state (quasistatic) equations are further utilized to determine the unperturbed state of the system prior to transient initiation. The system response to a complete loss-of-electric power leading to buoyancy-induced natural circulation is calculated and compared to parallel calculations using DEMO and SSC-L simulation models.

An application of response theory to steam venting during a loss of AC power transient

Abstract We have applied the theory of response to the loss of AC power transient for an LMFBR design to determine the ultimate loss of coolant inventory and the sensitivity of this figure with respect to the initial conditions and input parameters. Using a simple four region heat transfer model, the analysis shows that 3717 kg coolant are vented after feed water is lost and before venting stops. The sensitivity analysis reveals that this figure is strongly dependent on design parameters and system assumptions. The uncertainty in the lost inventory caused by the uncertainties and correlations in the input parameters and initial conditions is found to be 3464 kg. We thus report the result of the calculation as and conclude that the available inventory of 8775 kg is sufficient to ensure an adequate heat sink.