planar magnetic - goose1987/buck-CC-CV GitHub Wiki
Optimization through iteration
iterate over discrete core geometry and material to optimize loss based on desired max output current. Increase winding results in lower flux swings but higher DC loss. AC winding losses are dependent on geometry and are accounted for later in finite element modeling.
matlab scripts : https://github.com/goose1987/buck-CC-CV/tree/master/magnetics
EPCOS/TDK cores
N95 material gave lowest loss overall but not readily available. Go with N87 which is more common
Assume that space is at a premium so use EQ25 + PLT, N87 material even at the higher loss
Core : EQ25 Material : N87 Turns : 5 gap : 240 um (mechanical tolerance based on vendor, Elna magnetic can hit +/- 25um) Leff : 12 uH
finite element model
15A peak at 12A RMS
locate winding far from gap to minimize proximity effect