Mechanical Engineering Journal (Feb 2016)

Natural convection heat transfer from vertical rod bundles in liquid sodium

  • Koichi HATA,
  • Katsuya FUKUDA,
  • Tohru MIZUUCHI

DOI
https://doi.org/10.1299/mej.15-00596
Journal volume & issue
Vol. 3, no. 3
pp. 15-00596 – 15-00596

Abstract

Read online

Natural convection heat transfer from vertical rod bundles in liquid sodium was numerically analyzed for three types of the bundle geometry (two parallel, equilateral triangle and equilateral square arrays). The unsteady laminar three dimensional basic equations for natural convection heat transfer caused by a step heat flux were numerically solved until the solution reaches a steady-state. The PHOENICS code was used for the calculation considering the temperature dependence of thermo-physical properties concerned. The 2 to 4 test rods for diameter (D=7.6 mm), heated length (L=200 mm) and L/d (=26.32) were used in this work. The surface heat fluxes for each cylinder were equally given for a modified Rayleigh number, Rf,L, ranging from 3.06×104 to 3.14×107 (q=1×104 to 7×106 W/m2) in liquid temperature (TL=673.15 K). The values of S/D, which are ratio of the diameter of flow channel for bundle geometry to the rod diameter, for the rod bundle were ranged from 1.4 to 3 on each bundle geometry. The spatial distributions of local and average Nusselt numbers, Nuθ,z and (Nuav,B)N, on vertical rods of a bundle were clarified. The values of average Nusselt number, (Nuav,B)N,S/D, for three types of the bundle geometry with various values of S/D were calculated to examine the effect of the bundle geometry, S/D and Rf,L on heat transfer. The bundle geometry for the higher (Nuav,B)N value under the condition of S/D=constant was examined. The correlation for (Nuav,B)N,S/D for three types of bundle geometry above mentioned including the effects of Rf,L and S/D were developed. The correlations can describe the theoretical values of (Nuav,B)N,S/D for three types of the bundle geometry for S/D ranging from 1.4 to 3 within -7.44 to 10.73 % difference.

Keywords