Coupling parameter, E, and the total energy dissipated by the magnetosphere, UT, are determined for six disturbed periods, following three known criteria for UT computation. It is observed that UT exceeds E for Dst < -90 nT, for alI models. Differences between models reside on the estimated valnes for the particles' life time il1 the equatorial ring current. The values of TR, used in the models, are small during the main phase of the di."turbance, in disagreement with the charge exchange life time of the majority species, H+ and O'-. Based on this conclusion, a different criterion to calculate TR is proposed, differentiating the different stages of the perturbation. TR is calculated, for the main phase of the storm, from the rate of energy deposition estimation, Q, in the ring current. For Dst recovery phase, the vallles are obtained from a ring current decay law computation. The UTvu calculated, physically more coherent with the processes occurring during the event, is now smaller than expected. In this sense, it is understood that the power generated by the solar wind-magnetosphere dy- namo, should also be distributed in the inner magnetosphere, auroral zones and equatorial ring current, as in the outer magnetosphere, plasmoids in the tail shot in antisolar direction. A further adjustment of E, with the Chapman-Ferraro distance, 10' variable, has been made. Although the reslllts, improve the estimation of E, they are sti!l smaller than UT, except UTNU, for some disturbed periods. This result indicates the uncertainty in the computation of the input energy, by using the many expressions proposed in the literature, which are always presented as laws proportional to a given group of parameters, with an unknown factor of proportionality, which deserves more detailed physical analysis.