 |
|
|
|
I'm one of the guys that tried and failed, exactly for the reasons you mentioned. I always figured a flyback was easier to design and required fewer components. If anyone every gives you this argument, I know the counterargument. After you add the additional magnetics required to meet Part 15, the flyback has the same essential number of components. Not to mention the LIVING HELL I went through trying to get it quiet enough for Class B. I finally made it, but it took a six layer board, a very specialized common mode choke, and several different layout attempts. My successor went through SIX iterations before admiting that he couldn't do it either. A mutual acquaintace of both of ours finally solved it with a very improbable but effective two-layer (they were already at eight) board where each loop was replaced with a differential pair of traces. Aside: Don't every let anyone tell you that stripline traces have emission characteristics equal to differential pairs, they don't -- not unless you bring your system ground into the HV area, which will send the safety guys into orbit. So, I find myself needing a new offline converter for my current project and I am NEVER going to design in another flyback. I'll look over the guide at your website, thanks again. -J.Abel | When Not to Design With the Boostbuck Family of Co | | Isolation of the Cuk Converter | Historical Perspective | | Return Home | Safety First | Lesson in Ethics | The Consequences of Bad Design | Real World Applications | Deals | e-mail me! | |
||
