|
Home > Archive > Alternative Power sources > December 2005 > Re: scratch built wind generator - part#2 - posted
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Re: scratch built wind generator - part#2 - posted
|
|
|
|
| daestrom 2005-12-29, 6:21 pm |
|
"FrankG" <frank@theworkshop.ca> wrote in message
news:UemdnXxR1tqj-C7enZ2dnUVZ_tWdnZ2d@storm.ca...
> http://www.theworkshop.ca/energy/hdgen/2/hdgen2.htm
>
>
I don't mean to discourage you, but having worked in a motor/generator
repair shop, you still have a problem with your magnet/coil layout.
As I understand your design, you have 24 magnetic poles, and 8 coils. And
from your description, the magnets are arranged on one disk, and the coils
on another that is to be mounted very close to the magnet disk. And there
will be a steel/iron plate behind the coils? That part I can't tell for
sure.
Looking at how the magnets are arranged, N pole of one next to the S pole of
the next, I'm not clear how the magnetic flux will pass from the pole of
one, across a coil side, into the steel backing and return across another
coil side back to the S pole of the neighboring magnet. It seems the flux
would have an easier path to just jump from one magnet's N pole to the next
magnet's S pole and only a fringe of flux would pass up wards from the disk
to where the coil is spinning.
Instead of breaking the magnets in half, you might try mounting them so the
N pole of one is next to the N pole of the next. Followed by it's S pole
next to the S pole of the third, and so on. Then the flux from both N poles
would be opposing, and would have to rise up out of the disk, to where you
have the coils spinning. This would be hard to first hold in place since
the magnets would naturally repel one another, but I think you'll get more
voltage for a given coil setup/speed.
Another thought is that ideally the number of coils should equal the number
of poles in a single phase machine. With the odd ratio of 3 poles per coil,
however you connect the coils together, there will be times when the voltage
developed in one will be opposing that of another. Mind, I understand that
winding 24 coils to fit in the same circumference would seem to be a
challenge. But it is done all the time with commercial machines by
overlapping the coils. If one side of a coil is over a N pole, the other
side doesn't have to be over the adjacent S pole. It can be over a S pole a
couple of magnets away. If you try my idea about arranging the magnets so
poles oppose, then the coil can be wound sized so when one side is at the
'gap' between two magnets, the other side can be at the 'gap' between two
other magnets, some distance away. Then arrange all the coils so the left
side of each one tucks under the right side of its neighbor.
You can also 'flatten' the coils, spreading the conductors wider around the
circumference a bit. Don't spread them out so far that different turns in
the same side of a coil span more than about 3/4 the distance between N and
S pole pairs though. This spreading allows you to keep the 'thickness' of
the coils down so more copper is closer to the magnets and there will be
less gaps where there is no copper conductor.
If you ever do want to go three phase, you are going to need at least 3
coils for each magnetic pole, so finding ways to overlap / spread the coils
could be worth the effort.
Three phase will give you smoother output with less 'cogging' when you load
the unit. Or is this the reason you have so many more coils than poles, to
reduce the 'cogging' effect even though it sacrifices voltage generated??
daestrom
| |
| FrankG 2005-12-31, 12:21 pm |
| Daestrom,
Thanks for taking the time to provide such a thoughtful post...
I have had others offer similar thoughts, though I don't think anybody had
suggested alternating "Opposing" poles.
The next posting (later in January 06) will document a few variation on the
magnet placement and the results with the same coils as a reference point.
So far I have retrieved 48 Single pole or half magnets from the pile of HD
magnets and arranged them in a ring of 16 poles alternating N - S - N - S
.... The reuslts were consistently a lower output voltage for 1 layer, 2
layers and finally 3 layers .
With 3 magnets stacked over each other over 16 poles it equates to the same
quantity of magnetic material as 24 poles in 2 layers. The difference being
the spacing of approx 2/3rds of a pole gap between magnets.
My rational for using "Factory" shaped single pole magnets was that I had a
nagging suspicion that the act of breaking the dual pole magnets would
weaken then as a result of the mechanical force and that there would be a
measure of inconsistency in how accurately they parted along the division of
N & S.
Over the next span of time I have available, I will try your suggestion and
document accordingly, as well I feel that I can invest a bit more time with
just a pair of dual-faced magnets and the "Black-Sand" to more clearly
visualize the flux properties and how they are modified by adjacent magnets.
FrankG
www.theworkshop.ca
frank@theworkshop.ca
"daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote in message
news:a7Zsf.56303$XC4.37601@twister.nyroc.rr.com...
>
> "FrankG" <frank@theworkshop.ca> wrote in message
> news:UemdnXxR1tqj-C7enZ2dnUVZ_tWdnZ2d@storm.ca...
>
> I don't mean to discourage you, but having worked in a motor/generator
> repair shop, you still have a problem with your magnet/coil layout.
>
> As I understand your design, you have 24 magnetic poles, and 8 coils. And
> from your description, the magnets are arranged on one disk, and the coils
> on another that is to be mounted very close to the magnet disk. And there
> will be a steel/iron plate behind the coils? That part I can't tell for
> sure.
>
> Looking at how the magnets are arranged, N pole of one next to the S pole
of
> the next, I'm not clear how the magnetic flux will pass from the pole of
> one, across a coil side, into the steel backing and return across another
> coil side back to the S pole of the neighboring magnet. It seems the flux
> would have an easier path to just jump from one magnet's N pole to the
next
> magnet's S pole and only a fringe of flux would pass up wards from the
disk
> to where the coil is spinning.
>
> Instead of breaking the magnets in half, you might try mounting them so
the
> N pole of one is next to the N pole of the next. Followed by it's S pole
> next to the S pole of the third, and so on. Then the flux from both N
poles
> would be opposing, and would have to rise up out of the disk, to where you
> have the coils spinning. This would be hard to first hold in place since
> the magnets would naturally repel one another, but I think you'll get more
> voltage for a given coil setup/speed.
>
> Another thought is that ideally the number of coils should equal the
number
> of poles in a single phase machine. With the odd ratio of 3 poles per
coil,
> however you connect the coils together, there will be times when the
voltage
> developed in one will be opposing that of another. Mind, I understand
that
> winding 24 coils to fit in the same circumference would seem to be a
> challenge. But it is done all the time with commercial machines by
> overlapping the coils. If one side of a coil is over a N pole, the other
> side doesn't have to be over the adjacent S pole. It can be over a S pole
a
> couple of magnets away. If you try my idea about arranging the magnets so
> poles oppose, then the coil can be wound sized so when one side is at the
> 'gap' between two magnets, the other side can be at the 'gap' between two
> other magnets, some distance away. Then arrange all the coils so the left
> side of each one tucks under the right side of its neighbor.
>
> You can also 'flatten' the coils, spreading the conductors wider around
the
> circumference a bit. Don't spread them out so far that different turns in
> the same side of a coil span more than about 3/4 the distance between N
and
> S pole pairs though. This spreading allows you to keep the 'thickness' of
> the coils down so more copper is closer to the magnets and there will be
> less gaps where there is no copper conductor.
>
> If you ever do want to go three phase, you are going to need at least 3
> coils for each magnetic pole, so finding ways to overlap / spread the
coils
> could be worth the effort.
>
> Three phase will give you smoother output with less 'cogging' when you
load
> the unit. Or is this the reason you have so many more coils than poles,
to
> reduce the 'cogging' effect even though it sacrifices voltage generated??
>
> daestrom
>
>
|
|
|
|
|