How to calculate wire gauge for ferrite core transformer

[sbdada]

Hello, For general 50Hz Iron core transformer, I use the following SWG wire table to calculate wire gauge.




The 14th column of this table indicates current rating for 1000A per sq inch, and normally I use 2000A per sq inch. That means for 2A of current I use 20 SWG, and that's works fine.
But I don't know how to calculate wire gauge for high frequency ferrite core transformer. I observed they use thinner wire than normal iron core transformer even the current rating is lot more than the iron core. As example, the main transformer of an ATX PSU has about 22/20 SWG, 3 strands (don't know the exact gauge) which should carry only 6A according this chart, but they carry more than that.
So, how to calculate the gauge for this kind of high frequency transformer? And how many wire I've to use for a certain amount of current? Please help. Thanks in advance

[MagicSmoker]

Current density (commonly referred to as "J") for the windings in transformers is usually in the range of 2-6A/mm² (aka 200-600A/cm²). Now look at wire area rather than SWG or AWG or whatever.

[sbdada]

2-6A/mm², is that applicable for both ferrite core and iron core transformer? or only for ferrite core transformer? @MagicSmoker

[MagicSmoker]

2-6A/mm², is that applicable for both ferrite core and iron core transformer? or only for ferrite core transformer? @MagicSmoker

Technically speaking, it doesn't matter what the core material is; the optimum value of J depends entirely on the number of layers (ie - proximity losses), skin depth (ie - operating frequency) and desired temperature rise (in conjunction with core losses).

1000A/in² is equivalent to 1.55A/mm² which is excessively conservative for a mains transformer, and likely to be counterproductive in a HF switchmode transformer because proximity losses are proportional to the number of layers so your AC winding loss will take a step increase with every new layer, and note that it only takes a few turns on top of a lower layer to constitute a new layer.
 

[T3sl4co1l]

Higher current density is allowable because the transformer is physically smaller.

Higher by 2-4 times, depending.

However, you may find that, for very high currents, unusually heavy wire has to be used, or divided up, possibly into many fine strands (Litz cable).  This occurs because the current is not flowing in the entire cross section of the wire, at high frequency and deep within a winding.

Tim

[sbdada]

Tougher than I thought Need more study. Can you guys please suggest some books on SMPS specially emphasized on transformer winding section? Anyway, thank you everyone

[Siwastaja]

Tougher than I thought Need more study. Can you guys please suggest some books on SMPS specially emphasized on transformer winding section? Anyway, thank you everyone

You could start from here:
http://ludens.cl/Electron/Magnet.html and http://ludens.cl/Electron/trafos/trafos.html