In a geocentric kinematically rotating ecliptical coordinate system in geodesic motion through the deformed spacetime of the Sun, both the longitude of the ascending node Ω and the inclination I of an artificial satellite of the spinning Earth are affected by the post-Newtonian gravitoelectric De Sitter and gravitomagnetic Lense-Thirring effects. By choosing a circular orbit with I = Ω = 90 deg for a potential new spacecraft, which we propose to name ELXIS, it would be possible to measure each of the gravitomagnetic precessions separately at a percent level, or, perhaps, even better depending on the level of accuracy of the current and future global ocean tide models since the competing classical long-term perturbations on I, Ω due to the even and odd zonal harmonics J ℓ , ℓ = 2, 3, 4, . . . of the geopotential ideally vanish. Moreover, a suitable linear combination of I, Ω would be able to cancel out the solid and ocean tidal perturbations induced by the K 1 tide and, at the same time, enforce the geodetic precessions yielding a secular trend of −8.3 milliarcseconds per year, thus strengthening the goal of a ≃ 10 −5 test of the De Sitter effect recently proposed in the literature in the case of an equatorial coordinate system. Relatively mild departures ∆I = ∆Ω ≃ 0.01 − 0.1 deg from the ideal orbital configuration with I = Ω = 90 deg are allowed. Present-day levels of relative accuracy in testing the geodetic and the gravitomagnetic effects in the field of the Sun and the Earth, respectively, are 6.4 × 10 −3 (Lunar Laser Ranging) and 3 × 10 −3 (Gravity Probe B) for the De Sitter precessions, and 1.9 × 10 −1 for the Pugh-Schiff rates of change of gyroscopes (Gravity Probe B). Other tests of the Lense-Thirring effect with the LAGEOS type satellites are ongoing in the field of the Earth; their overall accuracy is currently debated.