Battery charger for wind turbines

[nibor wind]

Has anyone ever come on the market with a circuit that would take the rectified voltage from a home made wind generator and charge a battery with it.  Tying the output directly is far from optimal.  Until the back emf of the generator exceeds the battery voltage nothing is harvested from the wind.  Then the only reason that the generator does not run at a constant speed is do to the resistance in the system.  These resistances include the winding resistance of the generator, the wire resistance down the tower, and to a lesser effect the resistance of the rectifier and battery.  So with a higher wind speed the blades tend to stall and though away a lot of the energy.  A properly designed circuit would load the blades at a constant TSR for optimum performance.  The voltage of these generators are proportional to their rotational speed and the energy in the wind is proportional to its speed cubed.  This means that the ideal circuit would draw a current off the generator proportional to its voltage squared.  I can envision a circuit that would do just that and only have one control to  adjust the optimum point for a particular turbine.  Maybe there would be several different models for various size turbines.  An over voltage circuit could even be included to connect a dump load when needed.

I have not seen such a thing but I can sure appreciate their usefulness.

[wpowokal]

Not sure if you are asking or telling but there are devices available, i leave it to others to explain. From my own experience if there is enough wind I get amps if not no amount of fancy electronic is going to give me meaningful amps.

allan down under

[Flux]

You seem to have discovered the mppt controller. Yes it does work and recently such things have just about become commercially available.

Don't worry about the point below cut in, there is no power there and at best if you reduce the cut in speed you will only see a few watts. You are absolutely correct about the rest and it will give a much increased performance if you can keep the prop at constant tsr.

Read through this board, someone is testing the Classic controller with a 10ft axial flux mill and getting very good results. Can't think of his name at present.

Flux

[Flux]

Just to add a little more I have a machine running with a controller that does as you suggest. I sense alternator speed ( frequency) and derive a signal proportional to speed squared ( multiplier). I then track the current to be proportional to this speed squared signal. When you include the inevitable losses the input fairly closely follows speed cubed.

This is an analogue control but the idea could be done digitally. As it is at present it needs setting up to suit a particular machine but modern digital methods could let it learn a characteristic and adapt itself to a particular machine without any set up. I don't see much hope for a continuously tracking mppt controller as done for solar, wind is a bit tricky and with gusts and inertia and all it probably needs a base curve to work from even if it has to learn that curve to get it right.

Flux

[nibor wind]

Flux,

It was a good point to use the frequency instead of voltage to know the speed. I also agree that you would want a curve that can only be adjusted very slowly or it would certainly get faked out by a gust.  With todays cheep micros it would not be that hard to do all of this digitally.  I see controllers on the market such as the Power One PVI-6000 (I have used these myself) that include an inverter and can sort of do a job like this.  I also see there are a lot of controllers with MPPT on the market to charge batteries off solar.  The only controllers I see for wind are simple apply full generator output to battery until fully charged and then apply a dump.  By not allowing the generator to run at an optimum speed they throw a lot of power away and cause the blades to stall and make a lot more noise.  I would guess that this is more then a 30% energy penalty.  I would rather have a smart controller and pick up that energy.  Of course I would do this using modern switching technology.

Do you think there would be a market for such a controller at a cost somewhere around $.30 per watt or do you think most hobbyist would rather build a 1 KW turbine and throw away 300 watts?  

Michael