Please log in to comment.Log in
High-PFC 1.8KW AC/DC Flyback Converter - Mike D.

AC voltage-triggered flyback transformer - This simulates now, after the diode that was directly across the 60 Hz source was removed, and the op-amp was replaced with a comparator of similar performance. Since there was no feedback around the op-amp in the original version of this design, it simulates more robustly with a comparator.

This comparator model does not have a direct power supply connection, but it does use a 0V to 12V swing, corresponding to the simulation results that show the approximately 12V rail available in this circuit (net8).

This design uses the "More Speed" and the "Convergence Assist" options in the Advanced Options for simulation. This improves speed and robustness, of course!

Mike Donnelly's picture



Mike Donnelly

Joined November 8, 2013
Design added Tuesday, April 11, 2017 | 9:06 am PDT


Robherc's picture

Thank you Mike. The diode you removed was a (vestigal) placeholder for the other half of the control circuit (can't have a very high power factor when you're only feeding from 1/2 of a wave), but I can see where having a diode connected to a 120V voltage source could become quite a burden on the simulation engine (I don't even want to see how many amps it was calculating would pass through the diode).

As for the OpAmp/comparator swap, you're absolutely correct that the OpAmp was being used directly as a comparator. In this half of the circuit, the switch is perfect, and simulates well; the only reason I was using an OpAmp here is that your circuit builder doesn't seem to have an option for re-defining the Vcc/GND feeds to a Comparator, and for the other leg of the circuit (active when AC line falls below AC GND) I needed to be able to give the comparator VCC/GND feeds with a different ground reference point.In practice, these modified rail voltages will be absolutely necessary to avoid exceeding max. ratings for the comparator & the MOSFET it will be controlling. Without a way to modify the Vcc/GND to the comparator in simulation, is there an alternate method you can suggest for circumventing this problem, so that I could simulate the full circuit (still have a lot of control circuitry to add, which is what I was really hoping to use simulation to improve)?

Thank you again for your help with this.