We are looking at the SLUA517 App Note on Bridgeless Design. We are trying to understand this design. A couple of things seemed odd to me about what the paper describes.
- On page 7 in figure 8 they show a schematic where the mosfet for the line-side is connected to GDA and that current sense transformer is connected to CSA. Similarly, the neutral-side is connected to GDB and CSB. Most bridgeless designs are intended to be controlled by a single phase part, and both mosfets are just switched by the same signal. During each half cycle, one mosfet is already conducting through its body-diode so switching it has little effect, the other side is the only one that controls incoming current. It seems like you could still use just one phase to control the circuit shown on that page. And still use the UCC28070 to control two bridgeless phases, instead of using one UCC28070 for each phase.
- On page 4 in the text they indicate that you can use two smaller inductors. Presumably, they are saying that since each inductor is still active you only need half the inductance in each one to have the same total inductance. However it seems like the return path inductance is going to be useless. If the current is reducing the inductor is supposed to create a voltage difference across it to oppose that change. However as soon as the voltage difference exceeds 1V the current will just flow through the return diode instead.
I've looked at bridgeless designs before. This paper does a nice job of laying out most of the designs I've seen. On the second page of that document (p166) it shows the same two inductor design in figure 6. But the text indicates that only one inductor is active at a time so you need two, full-sized inductors. If they were half-sized each, they would cost half as much and be half as big, so doubling the count wouldn't really have much effect on anything. If they need to be two full-sized inductors that design doesn't make sense for us, since the inductor is probably the biggest, most expensive part on our board.
