Reducing leakage inductance by sandwiching primary between secondary halves?

[treez]

Hello,
We are doing an Active clamp forward converter, with vin=+48v, vout=-40v and f(sw)=250KHz, and Pout = 160W (schem and ltspice sim attached)
NS/NP=2.11
We have more and thicker secondary turns than primary, so we wish to reduce leakage inductance by sandwiching the primary between two halves of the secondary. Presumably its ok this way round?..we usually sandwich the sec between two hales of pri.

 

[chuckey]

As the primary magnetic path is in the core, it is better to have the windings as "flat" as possible. i.e. wind the secondary over the primary with a layer of mylar tape as insulation.

Frank

 

[treez]

yes but I am convinced the sandwich winding, AKA interleaved winding, means lower leakage.
Application note 18 from powerint.com explains this
Question is , is it ok to interleave primary between two secondary halves?

AN-18
https://www.power.com/sites/default/files/product-docs/an18.pdf

 

[FvM]

As the primary magnetic path is in the core, it is better to have the windings as "flat" as possible. i.e. wind the secondary over the primary with a layer of mylar tape as insulation.

A cylindrical windings arrangement is presumed, I believe.

Windings have however non-zero thickness which involves leakage inductance. Interleaving windings sections (either primary or secondary or both) reduces leakage. There's a detailed discussion in Snellings classical book Soft Ferrites, Properties and Applications in chapter 11.7 Leakage inductance.

The answer to the specific question in post #1 seems obvious to me. Why should exchanging secondary and primary matter? (Presuming equal total windings height)

 

[mtwieg]

Sure it works the same way, regardless of which side you refer to as primary or secondary. Only difference would how you connect the EMI and safety screens (if you even use them).