This paper describes a sensorless control for a high-performance permanent-magnet synchronous machine (PMSM) drive. A differential algebraic approach is used to estimate the position (and speed). The relationship thus obtained for analytical estimation of the position is based on the control voltages, the measured currents, and their derivatives. Unlike the other estimators like the extended Kalman filter (EKF) that is complex and computationally expensive, the differential algebraic estimator is simple with a 1-D model. Global convergence of the estimator is demonstrated, and experimental results show the effectiveness of the method and its robustness to current sensor fault.