I am investigating the 1/f noise behavior of the AD8606. Please look at figures 28 -- 30 in the datasheet, which show the voltage noise density (nV/rtHz) vs. frequency. There are three plots, each covers a different frequency range. Otherwise, the plots are taken for identical parameter values as far as I can tell.
Figures 28 and 29 make sense to me -- they asymptote to the same high-frequency value (about 6.7nV/rtHz). However, Figure 30 asymptotes to a different high-frequency value (15nV/rtHz), and the low end noise values don't match those in Figs 28, 29.
Any thoughts about how to interpret Figure 30?