On Fri, Feb 22, 2013 at 5:33 PM, RussellMc wrote: >> I have been trying to wind a high frequency transformer running at >> approximately 100kHz, in a push pull configuration with a square >> wave of about 25% duty cycle in a step up configuration. > > > General answer. Someone may offer specifics: > > Litz wire uses small conductors so that the skin effect uses the > majority of the copper in each conductor. You then use multiple > conductors as required. > > Power transmission is when wire is being used to carry power between > points - perhaps in a cable trough or secondary sheath. > > Chassis wiring is short distance inside an item of equipment. > Differences in Imax will affect temperature rise which may be > reguilated differently on long runs. So: which value do I need to use for calculation ? > A transformer has the wires adjacent and confined so heat is shared > and cannot erasily esscape - so you'd tend to go with the lower rating > - or maybe even lower again. If you are using an eg iron powder core > (which are commonly used for power handling at 10's to 100's of kHz > then they often use an organic binder which has a maximum temperatute > rating. Get them too hot and the binder starts to fail so the losses > rise so thjey get hotter so ... . Regenerative failure. > > SO: > > Decide what area per ampere you need designing conservatively. > Make that area up using eg 38 gauge strands. > Strands handle skin effect. > Area handles temperature effects. According to the wire gauge table 26AWG is sufficient for full penetration at 100kHz but http://www.litz-wire.com/New%20PDFs/Litz_Designing_Calculations.pdf states 38AWG is needed. !! ?? The more I search, the more confused I am getting. Came across this site: http://www.danyk.wz.cz/dcdc_en.html It seems the turns ratio doesn't apply at all here ?? V1/V2 =3D N1/N2 V1=3D325V V2=3D12V V1/V2=3D27.08 N1=3D80 N2=3D6 N1/N2=3D13.3 In many of the SMPS designs that I am finding that the turns ratio do not m= atch the voltage ratio ? Regards, Manu --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .