Photoinduced transient dipole experiments are used to measure the effective charge separation distance, which is equivalent to the photoinduced change in dipole moment divided by the electron charge of flexible electron-donor/acceptor systems, D-(CH2)(n)-A, where D is 4-N,N-dimethylaniline, A is 9-anthryl, and n = 3, 4. We find that the dipole moments increase strongly with solvent polarity. For the compound with n = 4 (DBA4), analysis of dipole signals indicates that the effective charge separation distances in toluene, 1,4-dioxane, ethyl acetate, tetrahydrofuran, dichloromethane, 1,2-dichloroethane, 2-methylpentanone-3, 3-pentanone, and benzonitrile are 2.2, 2.5, 4.5, 4.7, 5.5, 5.5, 4.8, and 6.3 angstrom, respectively. These values can be understood as the root-mean-square charge separation distance in the solutions of different solvents. We assume that the folded contact configuration has a separation distance of 3.5 angstrom, the extended, solvent-separated configuration has a separation distance of 8.0 angstrom, and that they are the only two stable species after electron-transfer quenching. The formation efficiencies of contact radical ion pairs (CRIPs) and solvent-separated radical ion pairs (SSRIPs) are estimated in different solvents. The results indicate that a significant fraction of the ion pairs exist as solvent-separated ion pairs when the dielectric constant of the solvent is larger than 10. These results indicate that electron-transfer quenching can indeed happen at large separations in polar solvents. They also reveal that there is a barrier for ion pairs formed at large separations, hindering collapse to a contact separation of around 3.5 angstrom.