Extracting only the available power...

[Externet]

Hi all.

When a wind generator is turning slowly due to lack of wind, its voltage output is usually not enough to -say charge a Pb 12V battery-.
On medium wind, loading its modest output, will decrease rpm and its corresponding voltage into also an inefficient mode, but perhaps capable of delivering some charging power.

What is the proper approach to automatically obtain an amount of power that will not decrease its rpm to stall/unuseable and inefficient levels, but still provide some juice in adverse conditions ?
Is it a variable input DC to constant DC converter? If so, say it is outputting 5 volts and the converter turns that into fixed 13,8V ; the battery acceptance status could equally overload the generation capability.

In other words, to a certain amount of wind, how to load it accordingly to sort of maintain useable rpm ? For a say 10W capable wind, how to keep the wind generator load under that figure even when the battery-to-be-charged would be happy or in need to suck ten times more ?

Miguel

[MrAl]

Hi there,


My first thoughts are to build a controller. The controller would measure speed of the rotor and output power and control the load current so that the power stayed at a maximum. The output of the generator goes into a switching regulator that outputs 14v or whatever is needed to charge the batteries. The power level is controllable via a control pin that lowers or raises the current getting to the batteries. That's about the same as for a good solar panel controller except instead of batteries they may use the actual power line coming into the building.

One way to work the controller is as follows...
Force an increment (or decrement) of the current draw and calculate the power.
Compare the new power level to the old power level.
If the new power level is higher, keep the new current level and keep incrementing or decrementing
If the new power level is lower, decrement the current draw and keep decrementing or incrementing.
(Note that if we were incrementing we keep incrementing, and if we were decrementing we keep decrementing, unless the power decreased at which point we switch from incrementing to decrementing or vice versa)

The increment is usually kept small, and there is also some way to recover from a very quick decrease in power such as if the wind stops abruptly, then a small time later starts again.

Despite the somewhat apparent complexity, these days this can be done with a single low cost microcontroller and some relatively low cost sensors. The only thing not mentioned above is what to do with the speed measurement...the idea there is to come up with some correlation between speed and power and keep an eye on the speed as well and incorporate that into the algorithm for perhaps a better responding controller. The speed measurement may not even be required though doing everything via measurement of the current and voltage.

Another more direct method would be to measure wind speed and use that to control the output power based on a known correlation between power and wind speed (i think it's a third order response, but it's been a while for me on this subject).

We could probably make up a spice model for this entire system and study it a bit better inside a spice simulator.

[k7elp60]

My thoughts are very close to MrAl's. That is to control the current of a regulated voltage charger.

[Dean Huster]

I was thinking that you should monitor the speed of the impeller and if it falls below a certain value, lighten the charge current and if below some lesser value than that, disable all charging. This same tachometer circuit could monitor for over-speed during gale wind conditions, applying a brake so you don't rip the impeller apart.

[Bob Scott]

There must be an optimum load where most mechanical power is transferred to the load, just like impedance matching in electronics. I don't know, but I suspect that most power will be transferred when the load slows down the prop to 70%
(that's SQRT(2)/2) of the freewheeling speed the prop would make for that existing wind speed. I'm guessing. You'd need a mechanical expert to tell you for sure.

In any case, you need to know both the wind speed and the prop speed and send the information from those sensors to a servo that can optimise the load current draw for any value of wind speed that might occur.

[MorseDude]

Yeah, I have seen circuit diagrams on the web in different formats for that. The controller you want is called a "Charge Controller" and it does exactly what it says. It keeps control of the charge to the battery. It doesn't matter how fast or slow a windturbine turns, the charge controller will make sure it either stops or starts the charge on demand. Any excess charge can be 'dumped'. I have seen folks using a regular waterheater heating element for the excess power, seems to work great. You can build your own charge controller if you wish. Just Google it. -MorseDude