The permanent magnet synchronous motor (PMSM) control poses a challenging task since the dynamics of the system is nonlinear, there are hard constraints, and computational costs should be kept as low as possible. One way how to satisfy all of the requirements is to use the model predictive control (MPC) with a small length of short horizon in the MPC criterion and then try to replace the remaining long-horizon part with an appropriate approximation. This paper develops the idea of the long-horizon part approximation by the expected time to reach the required state of the PMSM. The reason of using this expected time is properly justified here which simultaneously causes that the number of user-defined parameters is reduced to minimum. Then, the MPC controller for the PMSM utilizing this approximation is designed for continuous region of admissible control actions. The advantages of the designed control algorithm are demonstrated by numerical simulations.