concentrating flux

[artv]

Hi all, was searching about laminated cores ,and "concentrating flux" was mentioned,but no detail was given.So I searched that but no luck.I was trying to find out if the laminites of the core are bigger than the coils and magnets, does that draw flux away from the coils??   Probably not very clear   eg: my magnet face is 1 3/4" long by 7/8" wide, my laminatrd core is 2 3/4" wide, so 1" of unused core on the inside diameter of the mag rotor. The coils on the core still leave about a 1/2" of unused core. Does the unused part of the core draw flux away from linking the coils???  Also I tried to go to that FEMM site but couldn't make heads or tails of it.....I wanted to simulate how the flux would link a coil if you ,cut the E-core from a transformer in half, and put each half on each leg of the coil. Do away with the laminated core altogether.I'm guessing (hoping) this would work. Wouldn't this "concentrate the flux " to where it is of only any use???.........I hope I was clear enough.........thanks.........artv

[Flux]

Not very easy to follow this especially without drawings but I think I have some idea what you are on about.

To understand this properly you need to consider magnetic circuits. These consist of the magnet. an iron flux path and the air gap. it is the flux in the gap that matters. This gap flux depends on the gap area and gap length. If you make the area larger the flux density falls and vice versa. Your magnet can only supply a certain magnetising force to force the flux across the gap.

With flux return paths on the sort of alternators used here the gap area is basically the area of the magnet. The flux does fringe somewhat and for wide gaps the effective area is increased opposite the magnet but unless the gap is stupidly wide it makes very little difference. In your case having a larger core area than magnet doesn't really matter, the flux will pass across in an area similar to the magnet.

Using your transformer U cores will make a workable scheme as far as flux use goes but you are producing something that will inherently cog and you will have lots of trouble trying to minimise this. It is effectively the scheme used in simple motor cycle PMAs ( where cog doesn't matter) and it is also much the same as the F & P washing machine motor. This type of construction also results in a winding that will become reactance limited at higher currents. Yes it works but is not one of the best ways to build something suitable for wind power.

Flux

[kevbo]

Much of the classic literature on permanent magnet machines is now somewhat dated.  The fundamentals have not changed, but the texts make a lot of assumptions about available material properties which have changed over the years.  Most of the progress in this field is due to advances in three areas:  Stronger magnets, higher temperature insulation, and lower loss iron (metglass).   

Building Piggot style axial flux alternators with alnico magnets would have been a waste of time.  Ceramic magnets made it workable, and the new rare earth magnets make it work pretty sweet. 

Rare earth magnets allow you to get enough field strength to have the alternator work well without using iron.  Now with iron and RE magnets you might be able to use less copper, or cheaper magnets...but the cogging that comes with putting iron in the coils is an evil worth getting rid of even if it costs some efficiency.  It doesn't matter how efficient a wind turbine is when it is running if it isn't turning because it doesn't have enough wind to overcome the cogging.

The improved magnetic materials I mentioned in the first paragraph don't do anything to solve cogging, and truth be known they only ring another couple of percent efficiency out of otherwise good designs, and the lower the frequency, the less helpful they are.