Q2Electric Drives

(a) Major Advantages of Closed-Loop Control over Open-Loop Control in DC Drives

Automatic compensation for load disturbances: as discussed in detail elsewhere in this general subject area, an open-loop DC drive exhibits speed droop as load torque increases (due to the uncorrected armature-resistance voltage drop), whereas a closed-loop drive continuously measures actual speed and automatically adjusts the applied armature voltage to counteract this droop, maintaining the commanded reference speed essentially independent of load torque variations within the drive's rated capacity.

Superior steady-state accuracy: the integral action within a properly-designed PI (or PID) closed-loop controller drives the steady-state speed error to exactly zero (for a constant-speed reference and constant load torque), achieving far tighter speed regulation accuracy than open-loop control (which has no mechanism at all to correct for the inherent speed-load characteristic of the motor itself) can provide.

Faster, better-controlled dynamic response: a closed-loop control system can be specifically designed (through appropriate PI/PID gain tuning) to achieve a desired dynamic response speed and damping characteristic when the drive experiences a reference change or a disturbance, rather than being at the mercy of the motor and converter's own uncontrolled, open-loop dynamic response characteristics.

Inherent overcurrent/torque limiting protection: the inner current control loop (in a properly-designed cascaded control structure, as discussed elsewhere in this paper) inherently limits the maximum armature current the drive will ever command, protecting both the motor windings and the power-electronic converter from damage during transient conditions (such as a sudden, large speed-reference step change, or a sudden load-torque increase) that could otherwise cause a dangerously high current surge in a simple open-loop system.

Robustness to parameter variation and aging: since closed-loop control continuously measures and corrects for actual system behavior (rather than relying on a pre-calibrated, fixed open-loop relationship between control input and resulting speed), it remains accurate and effective even as motor parameters (armature resistance, in particular) change with temperature, or as the motor ages, whereas an open-loop control scheme calibrated for one specific set of motor parameters would drift in accuracy as those parameters change over time or with operating temperature.

(b) Design of a Closed-Loop DC Motor Control System Using a Chopper Circuit

A practical closed-loop DC motor control system using a chopper circuit follows the standard cascaded control architecture: an outer speed control loop and an inner current control loop, as described in detail (with an accompanying block diagram) elsewhere in this paper.

Sensors employed: speed feedback is obtained using either an analog DC tachogenerator (a small DC generator mechanically coupled to the motor shaft, producing an output voltage directly proportional to shaft speed) or, in a modern digital implementation, an incremental optical or magnetic encoder (producing a pulse train from which speed is computed digitally by counting pulses over a known time interval); current feedback is obtained using either a current-sense shunt resistor combined with a differential/isolation amplifier, or a Hall-effect current sensor (providing galvanically-isolated current measurement without the power loss of a series shunt resistor).

Control algorithm: the outer speed loop compares the reference speed command against the measured actual speed, with the resulting speed error processed by a speed PI controller to generate a current reference command; the inner current loop compares this current reference against the measured actual armature current, with the resulting current error processed by a (faster-responding) current PI controller to generate the final duty-cycle command applied to the chopper's gate/base drive circuit — this cascaded PI-PI control structure, along with appropriate current limiting (clamping the maximum allowed current reference from the speed loop to a safe value) and, in a digital implementation, appropriate anti-windup protection for the integral terms (preventing excessive integral accumulation during periods when the current reference is saturated at its limit), provides the complete practical control algorithm for a robust, well-performing closed-loop chopper-fed DC motor drive.

Bandwidth Separation Between the Two Loops

A critical design requirement of this cascaded structure is that the inner current loop must be tuned to respond significantly faster than the outer speed loop — typically by a factor of five to ten in closed-loop bandwidth, or equivalently in terms of the loops' respective time constants. This is because the outer speed loop's own stability and performance analysis implicitly assumes that the inner current loop can track its commanded current reference essentially instantaneously (i.e., that the current loop appears as a simple, fast, near-unity-gain block from the speed loop's point of view) — if the current loop were instead comparably slow to the speed loop, the two loops would interact dynamically in a much more complex way, making the overall cascade difficult to tune for stable, well-damped performance and risking oscillation or instability. The current loop achieves its required speed because it deals only with the motor's electrical time constant (La/Ra, typically a few to a few tens of milliseconds), which is inherently much faster than the mechanical time constant governing speed response (determined by the load/motor inertia and friction), so the natural separation between electrical and mechanical dynamics in a DC motor conveniently supports this required bandwidth separation between the two control loops.

Illustrative tuning approach: using the standard 'magnitude optimum' or 'symmetric optimum' tuning criteria commonly applied to such cascaded drive control loops, the current loop's PI gains are first selected (based on the known or measured armature resistance Ra and inductance La) so that the closed current loop behaves approximately as a first-order lag with a small, fast time constant; only once this inner loop's fast, well-damped response is established is the outer speed loop's PI controller tuned (based on the total system inertia J and any known load-torque disturbance characteristics), treating the now-fast inner current loop as an ideal, instantaneous current source from the speed loop's perspective. This sequential, inside-out tuning procedure is standard practice in industrial cascaded drive controllers and ensures that changes made while tuning the outer loop do not require re-tuning the already-configured inner loop.

Practical block-level implementation: in a digital microcontroller or DSP-based implementation, both PI loops are executed as discrete-time difference equations, with the current loop typically executed at a much higher sampling rate (often synchronized to the chopper's PWM switching frequency, e.g., every switching period) than the speed loop (which may run at a slower, fixed rate, e.g., every few milliseconds, since mechanical speed changes are inherently slower than electrical current changes) — this differing sample-rate structure mirrors, and further reinforces, the fundamental bandwidth-separation principle underlying the cascaded control architecture, and is standard practice in essentially all modern digitally-implemented industrial DC (and, with appropriate d-q axis extensions, AC vector-controlled) drive controllers.