Go to Otherpower.com Home Page Go to Forcefield Shopping Cart Go to Wondermagnet.com Home Page
Front Page - [Homebrewed Electricity-- (wind) (solar) (hydro) (steam) (controls) (storage) (mechanical)] - Classifieds - Site News
Everything - Newbies - [Remote Living-- (housing) (heat) (light) (water)] - Reviews - Diaries - Our Products
Estimating alternator coil turns


By Herr Snefl, Section Newbies
Posted on Thu Sep 18, 2008 at 08:56:38 PM MST
Alternator for driving a grid-connect invertor.

Hi,

I'm mid-way through building myself a wind-turbine. Once I've proven I can generate a useful amount of power I plan to buy a Windy Boy grid connect inverter. Trouble is I don't know how many turns to use in the alternator. There's loads of information about 12v battery charging systems but much less about driving inverters.

Background:

3.2m blades, TSR=6.
300mm diameter alternator. 2 discs, 12 magnets per disc.
Magnets are 2x1x0.5 inch N40.
Air gap currently 19mm.

I've wound a 10-turn test coil and taken some measurements with a 'scope:

Measured  Measured  Measured  Calc     Calc
Hz        Vpp       Vrms      RPM      Vpp/RPM/Turn
5.35      0.59      0.195     26.75    0.0022
6.15      0.60      0.201     30.75    0.0020
10.61     0.98      0.344     53.05    0.0018
11.86     1.12      0.381     59.30    0.0019
16.89     1.64      0.532     84.45    0.0019
18.00     1.72      0.587     90.00    0.0019
19.40     1.84      0.635     97.00    0.0019
23.30     2.16      0.737     116.50   0.0019
24.60     2.32      0.732     123.00   0.0019
26.80     2.60      0.904     134.00   0.0019
29.90     2.72      0.980     149.50   0.0018
33.80     3.24      1.110     169.00   0.0019
42        3.88      1.390     210.00   0.0018

I've tried to calculate the number of turns...

Say 0.0019 Volts peak-peak per turn per RPM.

Windy boy 1700 spec says:
Vmin    139
Vnom    180
Vmax    400

Turbine is 3.2m => 10m circumference, TSR=6, hence at windspeed 3.5m/s:

rpm = (3.5 * 6 / 10) * 60 = 126

Hence for Vmin at 126 rpm:

turns = 139/(0.0019*126*3) = 194

(Times three is because I'm assuming 3 coils in series per phase.)

What worries me is that Hugh Piggot suggests 70-90 turns for a 12v system so I'd expect to need ten times that number!

What am I missing? Resistive losses in the coil? Ripple after rectification?

All suggestions welcome!

Estimating alternator coil turns | 4 comments (4 topical)

Re: Estimating alternator coil turns (3.00 / 0) (#1)
by Flux on Thu Sep 18, 2008 at 11:07:07 PM MST

Firstly either your frequency or speed is wrong. I assume frequency is the thing you measured so that makes the speed wrong.

For a 12 pole machine speed works out at 10 times the frequency so your 42 Hz would be 420 rpm.

The test coil ought to occupy more or less the same position as the final coil. Using 10 turns you can only get more or less a point at some place over the area of the final coil so it may not be a good average of what you finally get. It may be reasonable for now but I would suggest one test coil the final size before you finally commit yourself.

Not sure of your magnet grades but for typical flux values with a 100 turn coil at 130 rpm I would expect dc to be about 30v.

The ac line voltage for this would be 30/1.4 = 21v

If star connected then the phase voltage would be 21/1.7 = 12v that is 3 coils in series so one coil about 4v rms.

You get 1/10 of that at something like 12Hz or 120 rpm so if you have N42 magnets or better then that ties up reasonably well ( assuming your frequency is right not rpm)

For battery charging your cut in speed of 126 is too low but in theory if the Windy Boy tracks the cube law it may be ok, but you may not see any output even if it locks on grid at 126 rpm.

From my figures of 21v at cut in, your coil would need to be 139/21 x 100 = 660 turns.

Sorry I cant think in volts/rpm/turn, just confuses me, I haven't checked those figures.

Flux



Re: Estimating alternator coil turns (3.00 / 0) (#2)
by Herr Snefl on Fri Sep 19, 2008 at 02:28:55 PM MST

Thanks Flux.

Yeah, I messed up the Hz->RPM conversion didn't I! I assumed each of the 12 magnets generates a full cycle as it passes the test coil. Actually the magnets alternate N-S so its one cycle of output per pair of magnets. Oops.

Your reply has helped me spot my second mistake. I've assumed my rectified DC will be the same as the Vpp. In actual fact it will be half that (or 1.4 times the RMS as you stated).

If I take my original calculation and double twice to correct my mistakes I get 776 turns which is not too far from your figure of 660, which gives me some confidence.

BTW The Windy Boy seems to approximate cube-law using two straight lines. There seem to be some user configurable parameters for adjusting the slopes.

My next job is to wind a 700 turn actual size test coil. Once I've worked out which gauge wire to use...

Thanks again,

Neil.


[ Parent ]



Re: Estimating alternator coil turns (3.00 / 0) (#3)
by methanolcat on Fri Sep 19, 2008 at 09:16:58 PM MST

Please post a picture of that coil, I would love to see it.

  How much current do you think it will put out? I hope you have huge magnets.

Matt



Re: Estimating alternator coil turns (3.00 / 0) (#4)
by Herr Snefl on Sun Sep 28, 2008 at 01:39:20 PM MST

Hi Matt, I've finally gotten around to winding my test coil. I was hoping to get 750 turns on but only managed 500. Picked too thick a wire in the hope of keeping the resistance down. Also its ended up 12mm thick, a couple of mm fatter then I intended. The white stuff on the coil is not icing sugar, its the paper I used to line the plywood former. This coil is giving me 32v at 125 RPM. I'll have 3 coils per phase hence 96 volts. I need 139 volts to get the Windy Boy to cut in. Guess I could use star configuration which would give me 166 volts. Trouble is I don't want to exceed 400v in high wind! Some pics...




BTW the current will be quite low because the voltage is relatively high. Cheap rectifiers for me :-)
Neil

[ Parent ]


Estimating alternator coil turns | 4 comments (4 topical)
Display: Sort:
Menu
· create account
· How to use the board
· FAQs
· search the board
· Google search the board

Login
Make a new account
Username:
Password:

Total Views
  82 Scoop users have viewed this posting.

Related Links
· magnet
· Also by Herr Snefl

Powered by Scoop
You must be a registered user to post here. It's easy and free, and the link is on the upper right side of your page.
All trademarks and copyrights on this page are owned by their respective companies. Postings are owned by the poster, but may be deleted or moved at the ADMIN's sole discretion. The Rest © 2009 Forcefield.
You can Email the board ADMIN here. PLEASE include the username you signed up with!