The influence of changing the tube and particle diameters simultaneously at fixed tube-to-particle diameter ratio N, on the pseudo-homogeneous heat transport parameters, was studied in experiments in wall-heated packed beds. The packings used were porous ceramic, nylon and steel spheres. Three columns of different diameters were used at a ratio of N = 8. Two sets of experiments were carried out, between which the data acquisition system was changed and several pieces of equipment refurbished. Experiments were also carried out with nylon spheres at N = 4, using four columns of different diameters. The effective radial thermal conductivities (lambda(r)) for each conductivity packing were in excellent agreement, for both the cases N = 8 and N = 4, thus supporting the hypothesis that lambda(r) should depend only on the ratio N and not on the tube and particle diameters individually. The apparent wall heat transfer coefficients (alpha(w)) were in very good agreement for all four columns for N = 4. For N = 8 the alpha(w) values in the largest and smallest columns for each packing were in good agreement. The alpha(w) values obtained from the intermediate diameter column were systematically higher in almost all cases. Comparisons with earlier experiments in the same column suggested that the values from the present work were too high and that alpha(w) also depends only on the ratio N of tube to particle diameters.