Two Switch Forward + PFC SMPS design in progress

[treez]

Anyway, like you can see, i gave it an np/ns of 10....
so you can choose 60/6, or 120/12, etc etc

choose whatever number of turns you like , as long
as
1...you have the right np/ns
2...you have whole numbers of turns for np and ns
3....You dont dissipate too much in the wires of the coils
4...you dont overheat the core with core losses.
5....you dont saturate the core
6...the turns fit on the core
7...you can terminate the turns to the former
8...mind out for skin effect

So you could use a layer of approx 0.5mm wire for the pri if choosing 60 turns...
and then you could have a few parallel strands for the sec...and spiral wind them across the bobbin so that the turns are evenly spread across the bobbin..
..this tends to increase coupling between pri and sec.

If your number of tunrs is low, then you often have ti use paralle strands to get the coil layer to fit the bobbin width...but as discussed , you can always do spiral winding.

 

[kkeeley]

Thanks again for all your help, I haven't yet had a real chance to fully study your examples, on first look they appear to make sense so far.

 

[treez]

The excel design sheet shows i considered it as a transformer isolated buck converter.....(which is what a 2tf is)
I took the condition of min vin and Continuosu conduction mode.
I then selected the max Duty cycle.
I then calculated what the "effective vin" would be, in order to achieve the vout with that "effective vin".
I then calculated what turns ratio would be needed to give me that "effective vin" from the actual vin.

...when i then did the transformer design, i ended up adjusting the max duty to suit the transformer design.......(i was juggling the amount and thickness of turns etc etc)

The overall equation for a 2tf is vout/vin = ND, where N = NS/NP (that of course assumes continuous conduction mode)

- - - Updated - - -

i think when you first do it....you just have to jump in, make a mess of it....and then start correcting it...or start over, and over.
Use the simulator to check your calculations.

 

[treez]

Hello KKeeley,
Please if you wish, send me your email address, and i will send you my SMPS course...it is via a google drive link
..it has lots of info on SMPS design, and i have distributed it widely.
andymassey22@gmail.com

 

[kkeeley]

I'm interested to see your course, Hopefully it will help me to better understand what I still need to resolve in my design.

I hope to try and spend a bit more time on my design over the Christmas break, I'm hoping to at least have part of the design working in spice soon.

I've sent you a PM with my address.