Marginally and transiently populated conformational states of biomolecules can play important functional roles in biochemical processes, It is of significant interest, therefore, to develop tools for characterizing the structural and dynamical properties of these excited states. One recent development has been the emergence of spin-state-selective relaxation dispersion methods for quantifying dipolar vector orientations in invisible excited-state conformers through measurement of residual dipolar couplings (RDCs). Particularly powerful are H-1(N)-N-15 RDCs that can be measured with high sensitivity on fractionally aligned, deuterated, uniformly N-15-labeled protein samples. Fractional alignment also produces nonzero H-1(N)-N-15 dipolar couplings. These can be problematic for the extraction of robust H-1(N)-N-15 RDC values, and hence amide bond vector orientations, in cases where the amide proton of interest and a proximal amide proton have small chemical shift differences and a significant H-1(N)-H-1(N) dipolar coupling. Here, we show that while this strong coupling effect leads to aberrant relaxation dispersion profiles, extracted excited-state H-1(N)-N-15 RDCs are for the most part only marginally affected. Experimental examples of such aberrant profiles are provided, as well as a theoretical consideration of the influence of this strong coupling effect and numerical simulations that assess its impact on extracted parameters.