Locally I can't find any cores bigger than 37mm OD 
which is greatly put a dampening to my next project in planning.
Someone said than a core like mine would do about 400w... Stacking is one option but I assume there is a limit to the amount of cores I can stack.
Another thing I've thought of is having a transformer per rail. So If I have to lots of stacked core with one per rail I could add a fair bit of power handling that way. The only problem is getting dual cores to run off one controller.
If I parralel them will it require me to extremly carefully match the impedence/inductance to that one core won't dominate?
I gather that a single secondary winding is required in only producing a single rail so I fit chunkier winding wire in there.
Another thread I started a while ago on core stacking someone said that by double the core u get a 30% decrease in windings, can anyone confirm this?
I know increasing the switching frequency decreases the size of the magnetics used but ppl have mentioned tradeoff between effiency and switching frequency. For this type of project what would be a suitable frquency?
Some project info -
Powering a Class D (or large Class AB if the Class D amp doesn't work out) in excess of 2kw bridged
+-44v rails using the Rod elliot design (the one with the bipolar stage and feedback)
Using SG3525 with 10 (5 per side) NTP75N06L MOSFETS
Current monitoring/protection by microcontroller (still in the works)
Standard 13.8v operating voltage
Thanks everyone
which is greatly put a dampening to my next project in planning. Someone said than a core like mine would do about 400w... Stacking is one option but I assume there is a limit to the amount of cores I can stack.
Another thing I've thought of is having a transformer per rail. So If I have to lots of stacked core with one per rail I could add a fair bit of power handling that way. The only problem is getting dual cores to run off one controller.
If I parralel them will it require me to extremly carefully match the impedence/inductance to that one core won't dominate?
I gather that a single secondary winding is required in only producing a single rail so I fit chunkier winding wire in there.
Another thread I started a while ago on core stacking someone said that by double the core u get a 30% decrease in windings, can anyone confirm this?
I know increasing the switching frequency decreases the size of the magnetics used but ppl have mentioned tradeoff between effiency and switching frequency. For this type of project what would be a suitable frquency?
Some project info -
Powering a Class D (or large Class AB if the Class D amp doesn't work out) in excess of 2kw bridged
+-44v rails using the Rod elliot design (the one with the bipolar stage and feedback)
Using SG3525 with 10 (5 per side) NTP75N06L MOSFETS
Current monitoring/protection by microcontroller (still in the works)
Standard 13.8v operating voltage
Thanks everyone