Fluid mixing studies in a hexagonal 61-pin, wire-wrapped rod bundle

Two wire-wrapped rod bundles with different leads (6 in. and 12 in.) were constructed with geometric parameters similar to proposed LMFBR fuel assemblies. Rod diameter was 0.25 in. and pitch-to-diameter ratio was 1.26. These two bundles were tested in a flow loop which was designed and built for mixing experiments. Fluid mixing was studied by means of salt tracer dispersion. Salt was injected at various radial and axial locations in the bundle via injection rods, and then the dispersed distribution was measured at the bundle exit by means of 126 specially designed electrical conductivity probes inserted into the bundle subchannels. The data collected showed a strong swirl flow around the bundle circumference and periodic variation with axial injection location. Data from turbulent runs was generally good with mass balances averaging 90% and having a spread of +- 25%. The laminar data collected was generally poor because of a ''striping'' phenomena and injection instabilities. Data were compared with calculations using the ENERGY computer code. The comparison between ENERGY calculations and the data was not good for laminar flow and was only fair in the turbulent cases. It was found that turbulent data could be best characterized by the ENERGY parameters C1 = 0.19 and epsilon1* = 0.025 when the lead was 6 inches; for a 12-inch lead the parameters were C1 = 0.16 and epsilon1* = 0.012. Pressure drop data was also taken from the two bundles and it too showed a periodic variation with axial location. Friction factors derived from the data were generally higher than predicted by available correlations. These data suggested that traditional flow split calculations could be in error and that the laminar-turbulent transition occurs over a broad Reynolds number range in wire-wrapped rod bundles