PM windgen questions

[Chris05]

Hey guys.

Im a new member of this website, its great. Ever since i saw thebackshed.com i have been hooked on this stuff. I have constructed a couple of small wind gens now but i want some more serious power.

Im in Australia, and i want to build a PM windgen to power some outside lights. I have a good knowledge of electronics and motors. I have made wound many custom motors for my remote control cars and stuff like that....

Anyhow... Ill cut to the chase.

My questions are relatively simple.... (at least i hope they are

1. Is there some sort of formula for calculating how many turns of wire you need on the stator to generate X amount of volts and Y amount of amps? Or is it just experience after a while of working with specific magnets and layouts you just know what to expect?
   
2. I have contacted a company called Aussiemagnets. They have given me a good deal on these magnets: LINK
   
   I know that they arn't exactly very big but its all i can really afford and find... I plan on doubling them up. Does anyone see any problems with them?

   
   Thanks alot guys... Your help is appreciated.
   
   Chris

[Flux]

How much do you want to produce? Those magnets are tiny, I don't know the value of the Ausie dollar but they seem very expensive.  #3510 50 x 25 x 12.7 is a much better bet but the price seems crazy and they do not even admit to the grade of neo.

Air gap alternators do obey the laws of nature and performance can be predicted quite well from basics, but it takes a bit of experience. If you throw in iron cores then there are variables that are not easy to predict. I advise you to work from a test coil for determining cut in speed. You can predict the amps into a battery from the resistance but there are again a few factors that you learn by experience.

The chances of doing it all from theory first time are not that good especially when you have magnets of unknown grade.

If you want much over 100W then those tiny magnets are a hard way to get it.

Flux

[seanchan00]

Hi Chris05,

I am Sean from Malaysia and although I am only a medical doctor with no background on anything electrical, I felt confidant enough to start my own PM wind turbine with an eight ft rotor. I however purchased 24 Neodymium 1" X 2" X 1/2" magnets from Otherpower supplies. They are 4 times the power of the mags you sourced mass for mass and only at US 6.50 if 20 and above are purchased. The link is here, see forself

http://www.wondermagnets.com/cgi-bin/edatcat/WMSstore.pl?user_action=detail&catalogno=0076

Strangely after all my reading in this subject, I too have the question that you asked not yet figured out. I am still reading more and trying to decide the size & number of turns to to best match the magnets I bought. I am using  two 9 inch brake discs to make a Dual magnet sided alternater with 12 poles on each disc and 9 coils in between.

Sure hope some old hands out there give a detail answer. Will be watching youe post laik a hawk.

Sean Chan.

[DanB]

I think you'd do very well to find some larger diameter rotors.  9" is really to small to make good use of your magnets.

If you can find some rotors at least 11" diameter, preferably 12" - then we could tell you exactly how many windings of what size wire you'll require.  Finding larger rotors to fit your hub shouldn't be too difficult - if you want to use brake rotors, go to a brake shop and look in their recycling dumpster.  Larger brake rotors should be easy to find...  the 9" ones you have now are really too small.

[kitno455]

nah, he's in a part of the world where cars dont have 12 inch rotors, unless he steals the rotors off his prime ministers Benz limo

best bet is going to be automotive flywheels or truck brake drum/auto steel wheel style. (or custom cut from flat stock)

allan

[Cinosh07]

Hello,

There is a copy of an e-mail i've receive from ed from windstuff website. Couldbe usefull for you.

Cinosh07

The following formula will get you in the ball park for how many turns of wire you'll need...

Voltage= NARBP*2

where:

N= number of turns

A= area of 1 magnet in square meters (  the triangle neo's are .001525 )

R= revolutions per second ( rpm/60 )

B= flux passing the coils in Tesla's ( gauss / 10000 )

P= total number of poles

To find number of turns:

Voltage / ( P * R * B * A)/2 = Number of turns

B is confusing because not every one has a Magnetometer ( gaussmeter)  so I'll give you what I've come up with while working with different versions:

On a dual rotor system, not exceeding an air gap of .75" between the magnets you'll see 0.5Tesla through the coils

A simple coil over laminant will get around 0.4 Tesla throught the coils

And a slotted silicon laminant with an air gap of .030 between the magnet and stator will see slightly over 1 Tesla through the coils with these magnets.

As an example my 8 inch dual rotor shown on my site, using 0.5T in the formula to find the amount of turns per coil would go like this...j

Number of turns = 14volts / ( 12poles * 3.33rps * 0.5 Tesla * .001525 area) / 2 = 229 turns

There are 36 coils in that arrangement total and each phase has 12 coils....   229/12 = 19 turns per coil

Since its wired in star you can divide the 19 by 1.732 which equals 11 turns per coil

If you were using a 9 coil arrangement ( 3 coils in series per phase)  then 229/3/1.732 = 44 turns per coil

That should get you in the ball park for designing an alternator that will work with the blade you want.   You want to use the largest size wire you can fit into the space allowed for maximum power.   Its best to keep the resistance in the wire as low as possible to achieve maximum output.

As far as the relationship of coils per magnets... On a single phase machine you would have 1 pole per 1 coil ( so if you have 12 magnets you would have 12 coils for single phase ).   The normal 3 phase is 4 magnets per 3 coils.   So you can have 8 magnets and 6 coils or 12 magnets and 9 coils... the magnets are always added in multiples of 4 and coils are added in multiples of 3.   Another way is 2 poles per 3 coils where 8 poles would have 12 coils or 12 poles would have 18 coils.   And finally you can overlap the coils using a single phase arrangement with 3 phases where each pole would have 3 coils so with 8 poles you would have 24 coils and 12 poles you would have 36 coils.   The 4pole/3 coil arrangement is the most popular because of its simplicity and the coils are laid out in a single layer.

Hope this helps

Have Fun

Ed

www.windstuffnow.com/main

[Flux]

Yes that's quite a convenient little formula. Remember that the flux densities quoted are for Ed's magnets so you will have to alter the Tesla to suit your magnets and air gap length.

The 2 x1 x 1/2" magnets will be nearer 0.7 T  with about 3/4" gap for dual rotor.

Flux

[scoraigwind]

I am not sure exactly what is meant by voltage in ed's equations.

Here are some useful equations for coil design

Average open circuit voltage (emf) for an alternator will be

Eave = 2*n*A B(rpm/60) volts

where Eave is average emf per phase

n is turns per coil (multiplied by number of coils connected in series)

A*B is flux in Webers

A is total area of poles in square metres

B is average flux density at poles in Tesla

(rpm/60) is revolutions per second

The peak voltage will be about 50% higher Epeak=1.56* Eave

If a three phase arrangement of coils is used then they are usually

connected Star (Wye) giving root(3)1.73 higher output volts.


If this is fed through a rectifier, then the output is approximately

the peak with the forward biassing voltage of the diodes subtracted


Edc = Epeak*root(3)-1.4 volts for silicon diodes.


In this way you can establish the cut in speed of an alternator.

Where Edc = Vbat=battery voltage, the rpm can be found.


rpm = (Vbat+1.4)60/(1.73*1.56*2*n*A *B)


You can also use E to predict the output current.  There is no

current until Edc>Vbat.  After that, current I will be


Idc(Edc-Vbat)/R amps neglecting reactive effects (self induction)

Where the coils are wound on laminations, the inductance is strong

and it is hard to predict output current.  The current will be

limited by the inductive reactance at higher speeds.  With air-core

coils in a stator which does not contain iron, we can look at the

resistance R in ohms as the main impedance.

R =L/Aw*0.022(1+0.004*(temp-70)) ohms

where L is the length of copper wire in the coils of one phase in metres

Aw is the wire cross sectional area in mm (pi()*square(diameter)/4)

temp is wire operating temperature in degrees C

Copper Loss = square(Idc)*R watts

[kenputer]

Couldn't have said it better myself,ain't it great that we know so much and our exsplaination are so easy to understand! What I mean is I'm lost with that one:-(

[Chris05]

Hey, i know those magnets are pretty small

Its hard to find stuff within Australia at a reasonable price, i usually resort to overseas buying for stuff like this which i might end up doing.

Those formula's are great, accept i dont have a magnetometer... My dad might have one, ill ask him when he gets home.

Thanks for your help guys.

[Flux]

Good point, Ed's formula should give ac average volts per phase, for rms the factor should be 2.2.

I see that in his example he is treating it as though it predicts the dc or peak ac volts.

Perhaps he has found it gives the right answers with what he thinks is the right flux value.

Flux

[seanchan00]

Hi Senior members,

The advice by DanB to get a bigger brake disc is good to get the optimum use of my 1 by 2 by 1/2 inch Neo mags and I can easily do it. In Fact I should be using 9 to 10 foot rotors to get the full use of the 24 magnets. The problem with my requirements is that I am using the wind turbine in my 6 hectar durian orchard where I am an absentee part time hobby fruit farmer. My power requirements are to run by timer one 20 V fluorescence light and play a car mp3 player with double power amplifier for 12 hrs a day to produce louder sounds for farm use. My problem is to try to balance my immediate use needs without burning out the system since there is no one around to stop the turbine in case of freak storms and winds, therefore the choice of only 7 feet rotor. I am thinking of making the tail light so it will furl earlier to balance my recharge needs and to shunt the smaller extra to unneeded heating coils. In future when there my power needs increase I can make the tail heavier or make a bigger rotor. I will probably start with a wider air gap between the two rotors then adjust closer if not enough. However I will make the stator as thin as possible.

I can understand the input by Cinosh07 and I am much clearer now. I will use the thickest coil that can fit the unit so the heat produced is lower in case of freak winds to insure against burn out. I am also thinking of a mechanical brake powered by wind force so it will come to my rescue when there is super wind blowing.

The highly technical dissertation by our Guru Hugh of course is a very valuable reference but with my zero electrical background I gave up trying to digest it halfway. I have saved it in my computer for future reference when my knowledge increase enough.

SeanChan

New junior.

[seanchan00]

Hi DanB,

Thanks for your advice. After some deep thinking I have decided to cut two 3/8 inch steel plates to hold the 24 magnets. To cut the chase short just tell me for my 7 to 8 feet rotor :

What diameter to cut the steel plates.

The diameter of the stator I should make.

Size and turns of the 9 coils in the stator.

and I will follow from there. I have been reading all your Otherpower projects for almost a year

Thanking you in anticipation.

Sean Chan.

seanchan00@lycos.com

PS- How do I include my biodata and e-mail when I post in the Discussion board?

[domwild]

Chris,

Not wishing to stop you from making your own but you would have seen the Fisher & Paykel idea on the backshed's and other web sites.

Look for an abandonded washer during the kerbside pickup days and disassemble one. Should give you 300 - 500W or so. Less fun than creating your own!

Good luck!

[Chris05]

Yeah that was my back up plan mate..... Well it was my first plan but once again they are hard to find at the rice price in melbourne, just like everything is!

But i think i have sourced a cheap F&P smart drive setup.

Thanks for your help guys.