> The Axial Flux in this very descriptive
> http://www.windstuffnow.com/turbine%20kit.pdf has 12 
> magnet pairs for 6 pole
> changes per turn,
> It puts out 3 phase current from 3 coil sets of 3 each (9 
> coils)

> It has problem that some consider a solution, namely to 
> minimum startup wind
> speed, it doesn't use cores in the coils to better close 
> the magnetic
> circuits...
>
> The result is that the maximum achievable magnetic flux is 
> limited by the
> width of the air gap...
> No professional made motor or generator allows itself such 
> a 'fo pa'!

Arguable.
Engineering is about both optimisation and compromises.
If low wind speed startup and generation is adequately 
important to you and you can achieve the requisite power 
levels at an acceptable $ cost by eg using more expensive 
magnetics then if that is an acceptable tradeoff for you 
then it is an acceptable solution. Adding ferromagnetic 
material (iron, ferrite, ...) to the magnetic path can help 
you maximise the utilisation of both magnetics and 
conductors (copper, aluminum, ...) at the expense of 
increased weight and new complicating factors such as 
saliency (aka cogging), 'iron' saturation etc.

> We recently started this group
> http://tech.groups.yahoo.com/group/Performance_Axial_Flux/ 
> to develop
> functional improved Axial Flux Motors and Alternators!

I'll look at that with interest, but

> We know that cogging as produced by only 3 phases closing 
> magnetic circuits
> over the paramagnetic cores of only 3 phases is an enemy 
> to low wind speed
> wind generator startup...

While many 3 phase machines exhibit 'cogging', it is 
possible to produce three phase and even one phase designs 
that have essentially zero 'cogging', even though most 
commercial designs that use ferromagnetic material don't 
manage. One reason for this is that the target application 
is usually more sensitive to other factors where the cogging 
is less important than other factors such as maximum 
material and/or spatial utilisation.

This alternator

        http://others.servebeer.com/misc/alternator.jpg

has NO discernible cogging. It, and others like it, will 
turn with the slightest of input. Note that the numbers on 
the magnets are meaningless (even though the person who 
wrote them didn't realise this) and the magnets are 
magnetised en situ after insertion so the magnetic field 
pattern is not aligned with the physical magnets.

This alternator is capable of about 300 Watts at 1800 RPM 
but the person who designed this is about to produce a 1 kW 
or better unit at the same or lower RPM. He is not willing 
to be directly contacted by email as the world beats a very 
well worn rut to his doorstep whenever they discover him due 
to his genetal expertise in this area BUT if people are 
strongly interested I would pass on well framed queries.

Note that most alternators are NOT operated with sinusoidal 
waveforms but trapezoidal (quite close to square waves), 
with the voltages rapidly transiting in polarity at the zero 
crossings. This has major implications for how power is 
extracted.

As a generalisation, cogging/saliency is minimised when the 
net magnetic forces are balanced between all poles in a 
system as the rotor turns so that as a metal/magnet gap 
decreases at one point and equivalent gape is opening 
elsewhere. It is not overly difficult [tm] to arrange things 
so that net force is notionally zero. It's then a simple 
matter of engineering [tm][again] to try and make practice 
match notionality as well as possible.

Things like magnetic pole skewing can help. Gargoyle 
probably knows.


        Russell McMahon






-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist