here's another one option..see the forum link below..
http://www.diyaudio.com/forums/power-supplies/187973-1000w-smps-based-ir2110.html
http://www.diyaudio.com/forums/power-supplies/187973-1000w-smps-based-ir2110.html
would a litz alternative be multiple parallel windings of 0.25mm, or 0.3mm, or 0.4mm diameter wire?
I have opened up a few SMPS of various types from ~10VA to 600VA and none used litz wire.
ALL used paralleled copper wires.
It depends, like always if you can parallell wires.
You need to insure that the wires are subjected to a identical magnetic field along the entire winding lenght , otherwise there will be difference voltage induced in each wire and internal circulation of currents, causing heating.
Compare to parallelling power supplies without balancing voltage/output impedance.
One method is to twist the wires along the winding. Litz is weaved/twisted.
But it has other disadvantagess like difficult to solderconnect, expensive and occupies a lot of space.
Another method is to wind the wires parallell at the same distance from the core, assuming the field is homogenous buth then you run out of winding width quickly, and needs to add another layer to fit enough turns.
It is a matter of wire diameter, skin depth and most importand the number of winding layers. Read up on proximity effect, Dowell curves.
Texas Instrument has some very good app notes.
Google for Slup125, Slup126 (pdf files)+ Dixon.
They have an entire set of apps describing design
Edit: found the link to the collection at TIs
http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM401014
You need to insure that the wires are subjected to a identical magnetic field along the entire winding lenght , otherwise there will be difference voltage induced in each wire and internal circulation of currents, causing heating.
Compare to parallelling power supplies without balancing voltage/output impedance.
One method is to twist the wires along the winding. Litz is weaved/twisted.
But it has other disadvantagess like difficult to solderconnect, expensive and occupies a lot of space.
Another method is to wind the wires parallell at the same distance from the core, assuming the field is homogenous buth then you run out of winding width quickly, and needs to add another layer to fit enough turns.
It is a matter of wire diameter, skin depth and most importand the number of winding layers. Read up on proximity effect, Dowell curves.
Texas Instrument has some very good app notes.
Google for Slup125, Slup126 (pdf files)+ Dixon.
They have an entire set of apps describing design
Edit: found the link to the collection at TIs
http://focus.ti.com/docs/training/catalog/events/event.jhtml?sku=SEM401014
Last edited:
It's just ok leeno...just always have time to read and explore this forum...you can really learn more like me..
I think one good principle is to try it first to realize it..whether it is about litz, core material, size of wire, number of turns, etc..because this forum has a great number of opinions majority of which are from experience...it just depends on anyone which of those will be followed..the most important is there is learning after trying whether the result is good or bad..But always keep in mind that death is permanent but carefulness can overcome it...
I think one good principle is to try it first to realize it..whether it is about litz, core material, size of wire, number of turns, etc..because this forum has a great number of opinions majority of which are from experience...it just depends on anyone which of those will be followed..the most important is there is learning after trying whether the result is good or bad..But always keep in mind that death is permanent but carefulness can overcome it...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.