The design and practical application of digital control techniques to a solid-state DC drive system are described. The aim is to design a proportional-plus-integral optimal controller on the basis of an accurate linear model of the thyristorized drive system. This controller overcomes the steady-state error problem and provides perfect tracking of the input disturbances. The drive system consists of a converter bridge feeding a separately excited DC motor. A linear model of the drive system is identified using the unbiased recursive least squares identification approach. This model overcomes the inaccuracy of the analytical linear model for both the converter bridge and DC motor. Approaches to practical identification and digital control of the drive system are described. Experimental results are presented to illustrate the effectiveness of the identified model and the improved controlled system performance. In addition, a programmable starting technique of the drive system is tested.