Problem with RCD circuit of flyback converter

[mtwieg]

When I calculate MOS dissipation I use Irms^2xRds(on) + switching losses

Sure, though the switching losses can be tricky to calculate correctly, depending on what kind of accuracy you want.

As I remember, diode losses are calculated with Vf x Iav (forward voltage drop times average current, in my case Iav=1A)

Yes, this works fairly well for diode losses. Keep in mind that diode reverse recovery causes additional losses in the MOSFET or snubber network.

 

[loxly]

Oh, it doesn't have to be that accurate. It's mostly for the purpose to see if cooler is needed or not. We already have a good cooler for mosfet, as far as I can see now it won't be needed for rectifier diode

 

[loxly]

I have one small doubt regarding to flyback core air gap

As I live in a crappy country, I could only get RM 14 core with no air gap, so I have to make my own.
My calculated air gap is about 0.2 mm.
So I've put spacers (paper) in each leg (0.1 mm each gap). Which gap length should I achieve in every leg to effectively get the same thing as in 0.2 mm gaped center leg only?

The problem is I can't find much on it, and I've found two different information.

1. The gap length of every leg should be half of the gap in central leg gap implementation

2. The gap length of every leg should be 2/3 of the gap in central leg gap implementation

I must say I'm more into the second option. It has more sense, because only the half of the center leg flux is flowing through each outer legs. Magnetic reluctance of gaps is also the same as in gap reluctance in central leg (2/3 of central reluctance+(2/3)/2 parallel reluctance of outer legs)

Anyone know which is the correct approach?

I'm going crazy here

 

[FvM]

It's clearly the first one, presuming a core cross section that is constant along the magnetical path, which is more or less the case according to simple considerations of efficient core shape design.

 

[loxly]

Can you give me a simple explanation why, please?

...because the second one has more sense to me. Anyway I believe you're right, I'm still new in magnetics

I don't know if it is important in this case, but just to mention, both my primary and secondary windings are placed on central leg

 

[FvM]

both my primary and secondary windings are placed on central leg

Yes that's clear, otherwise the device won't work as a regular transformer.

The magnetical path forms a closed loop through the center of the windings and back on the outside. On the outside it is split into two halves that are magnetically parallel connected. The total path kength is the lenght of one of these pathes, the total air gap the sum of the center and one of the outer gaps.

 

[loxly]

Now I undrestand, thank you very much