The result of such dynamic variation is shown in Figure 18: at low input line the switching frequency does not

change (or changes slightly) compared to the case in which dynamic blanking time is not implemented, while

decreases more and more as the input voltage increases.

Figure 18. Typical normalized switching frequency variation over input voltage range

The function beneficially impacts also on the blanking time profile variation versus FB voltage, as shown in

Figure 19: depending on the input voltage, the dynamic blanking time reduces the maximum permitted switching

frequency also at medium/light load, thus further minimizing the related losses.

This method reduces the frequency variation span and, consequently, increases the peak of the primary side

current. The Line Feed-forward compensation must be adjusted to guarantee the same power capability at any

line condition.

Dynamic blanking time and valley synchronization (see relevant section) functions can be disabled connecting the

pin TB to ground. The decision if the function is used or not is taken at power-up: before the pin VDD reaches the

VDD-ON threshold, the current ITB-DIS (200 µA typ.) is sourced from the pin. If the TB pin voltage is lower than the

VTB-DIS threshold (100 mV typ.), the pin is assumed to be connected to GND, the related circuitry is disabled, and

the information is latched until the next power-on.

Figure 19. Blanking time variation vs. FB voltage with dynamical Blanking time active

5.8

Valley synchronization

The main feature of the QR converters is that the turn-on occurs at the valley of the resonance after the

secondary demagnetization, which is a function of primary inductance, LP and total capacitance of the drain node,

CD. With respect to the transformers’ demagnetization instant, the first valley occurs with a delay given by the

formula:

VIPERGAN50

Valley synchronization

DS13888 - Rev 2

page 19/37