Controller IC and one-shot form resonant controller
Edited by Bill Travis
C Rajan, Centre for Advanced Technology, Indore, India -- EDN, November 28, 2002
Resonant power supplies are popular because of high efficiency, low noise, and compactness. You can implement a resonant buck or boost converter using a single switch. The regulation of the output in such a converter derives from using a constant on or off time and a variable frequency. The UC1864 controller IC (Texas Instruments, www.ti.com) meets all the requirements for a single-switch quasiresonant converter. But you also can achieve this performance with the inexpensive and popular UC3842 current-mode controller in conjunction with a one-shot multivibrator. This circuit gives the desired current-mode operation along with constant off-time, variable-frequency operation. Figure 1 shows the controller circuit, configured as a resonant flyback converter for high-voltage generation. For the implementation of this function, the current-mode controller should have 100% duty cycle; thus, you can use the UC3842 or 3843. This circuit uses the 3842. The operation of the circuit is as follows.
The output of IC1, the UC3842, is in a high state until the internal current-sense comparator goes high, and then the output of the IC switches low. As the output voltage goes low, IC2, the negative-edge-triggered one-shot CD4098B, triggers, and it generates a pulse. The pulse duration is a function of the values of R5 and C3, and this time interval is the constant off-time of the UC3842. The 
output of the one-shot sinks the base current of transistor Q1 during this period, and so the transistor conducts. This action adds a 5V pulse to the ramp at the RT/CT pin (Pin 4). This voltage remains at 5V until the pulse ends and goes high. As it goes high, Q1 turns off, but the voltage across CT is 5V, which is higher than the upper threshold of the RT/CT pin, so the internal circuitry pulls the pin's voltage to zero, and a fresh ramp and a fresh cycle start.
You must select RT and CT in such a way that the output voltage goes low before the ramp voltage reaches its upper-threshold level. If not, a glitch can occur in the output, arising from the time constant consisting of the output resistance of Q2 and CT. Figure 2 shows the voltage waveforms at low and high duty cycles. Traces D and C show the output and RT/CT-pin voltages, respectively, at a low duty cycle and, hence, high frequency. Traces B and A show the same voltages, but at a higher duty cycle (low frequency). You can see that the off-time is fixed in both conditions, and only the frequency changes to achieve the desired duty cycle. The voltage at the RT/CT pin gives a clear view of how the circuit modifies the ramp. This control circuit generates a 2-kV, 5-mA dc power supply using a resonant flyback scheme, and it performs well.
Is this the best Design Idea in this issue? Select at www.edn.com.
