In this paper we present results of a global gravity field recovery using half a year of CHAMP data. We use the energy integral of the motion of a satellite to transform satellite velocities into values of gravitational potential. The feasibility of this approach has already been demonstrated by several groups, using CHAMP reduced-dynamic orbits. We show, that the potential recovered from this kind of orbits depends on the a priori gravity field used for orbit determination. Thus, it cannot be excluded that errors present in the prior field propagate into the new CHAMP gravity model. It is the intention of this paper to avoid this dependency through the use of kinematic orbits, which are free from prior information. The derived potential model, TUM-1S, is validated by comparison to ground data and by satellite orbit residuals. It is shown to be comparable in quality to other state-of-the-art gravity field models.