The usual control objectives in a high-power-factor rectifier
are to keep the low-frequency portion of the input current
proportional to the input (full-wave-rectified ac line) voltage
and to keep the output dc voltage at a specified reference
level .
A general block diagram of the NLC controller for a
switching converter in the rectifier application is shown in
Fig. 1. At the beginning of a switching period, a short clock
pulse sets the flip-flop (FF), which turns on the power switch
. The control is based on sensing the switch current and
comparing the signal derived from the switch current with
a periodic nonlinear-carrier waveform . A signal
proportional to the cycle-by-cycle integral of the switch current
is used in the block diagram of Fig. 1. As an alternative, a
signal proportional to the instantaneous switch current can
also be used [6], [7]. At , and the
comparator output goes high and resets the FF, thus turning
off the power switch . The process is repeated in each
switching period. The NLC control can be viewed as currentmode
programming, or as one-cycle (charge) control, with a
judiciously selected nonlinear compensation ramp.
The periodic nonlinear waveform can be constructed
so that the input current automatically follows the input
voltage , as required for unity power factor rectification.
The waveshape depends on the converter topology and
type of the signal derived from the sensed switch current.
Signal affects the amplitude of the waveform and
determines the emulated resistance of the rectifier. As
shown in Fig. 1, is usually obtained at the output of an
error amplifier in the slow voltage-regulating loop. As opposed
to other control methods, the NLC control requires only switch
current and output voltage sensing and achieves simplicity
comparable to operation in the DCM. Sensing of the input-