Electric propulsion system for traction vehicles with automatic retard speed regulation

In a traction vehicle propulsion system comprising a pair of adjustable speed d-c electric motors which can either propel or dynamically retard the vehicle, each of said motors having armature windings adapted to be connected to a source of d-c voltage of variable average magnitude and having field windings adapted to be connect to a separately controllable source of excitation current the average magnitude of which is variable and depends on the value of a control signal that is supplied to said controllable source, said propulsion system including command means for producing a predetermined command signal when dynamic retarding of the vehicle is desired, the improvement comprising:

• a. first means coupled to both of said motors for providing a feedback signal representative of the actual rotational speed of the slower motor;
• b. second means for providing a speed reference signal of preselectable value;
• and
• c. control means connected to said first and second means and to said command means and responsive to the excitation current in said field windings for producing said control signal, said control means including
• (i) third means effective in the absence of said command signal for making the value of said control signal representative of any deviation of the absolute magnitude of said excitation current from a first reference magnitude which varies as a predetermined function of the magnitude of current in the amature windings of said motors, and
• (ii) fourth means effective when said command signal is produced and responsive to the rotational speed of the slower motor exceeding a first predetermined desired speed for making said control signal value representative of any deviation of the absolute magnitude of said excitation current from a second reference magnitude which is normally determined by the error between said slower motor speed and said first predetermined desired speed, whereby said excitation current magnitude is varied as necessary to minimize said error.