Modeling heat generation and flow in the Advanced Neutron Source Corrosion Test Loop specimen

Abstract

A finite difference computer code HEATING5 was used to model heat generation and flow in a typical experiment envisioned for the Advanced Neutron Source Corrosion Test Loop. The electrical resistivity and thermal conductivity of the test specimen were allowed to vary with local temperature, and the corrosion layer thickness was assigned along the length of the specimen in the manner predicted by the Griess Correlation. The computer solved the two-dimensional transport problem for a given total power dissipated in the specimen and stipulated coolant temperatures and water-side heat-transfer coefficients. The computed specimen temperatures were compared with those calculated on the basis of approximate analytical equations involving the total power dissipation and the assignment of the physical properties based on temperatures at single axial points on the specimen. The comparisons indicate that when temperature variations are large along the axis of the specimen, the variation in local heat flux should not be overlooked when using approximate equations or models. The approximate equations are most accurate near the center of the specimen where the heat flux remains closest to the average value, and in that region the calculated quantities agree closely with the results of the computer code. 4 figs., 1 tab.