The big wind farm fans drive a three phase induction (squirrel cage) motor. When the wind is too slow they are normally taken off line and the = blades are feathered. As wind rises, the blades are positioned to bring = the generator up to synchronous speed. Once the system is at = synchronous speed it is connected to the grid. As it is synchronous = with the grid it neither draws nor provides power. The blades are then = adjusted to drive the "motor" faster than synchronous. The difference = between the two speeds (the slip frequency) determines how much power is = delivered to the grid. Several conditions must be true in order for this to work. The base = load plant must be providing excitation. The wind farm generators = should not be providing most of the load. The grid's frequency isn't really "rock steady" it just looks that way. = A great deal of effort is expended keeping its average frequency solid. = You are correct though in suggesting that the wind farm generators don't = (or shouldn't) contribute enough to seriously perturb the overall = system. Phases are locked relative to each other by the design of every = generator in the system. Anybody who varied from the ideal would = quickly burn out their generator. The only affect of different angular rates from different generators in = the farm would be differences in their contribution of power. There is = no need to match up angular positions as the motors themselves will get = their excitation from the grid. Maintaining synch with the grid is the = easy part, it's essentially automatic. What's hard is ensuring that you = don't provide too much power (burn out your motor or get pushed over in = a strong wind), pointing into the wind, protecting yourself from gusts = and other not so obvious side issues. Dave Debbie wrote: >--- Howard Winter wrote: > > = > >>Debbie, >>There are a number of ways from adjusting the speed (field coils control >>that), using rotary converters (a motor driving a generator), to converti= ng = >>to DC and back. If you have great speed variation then the DC route is >>probably the easiest. >> >>Howard Winter >>St.Albans, England >> = >> > >Thanks guys! > >So .... OK when the wind drops and the blades are turning slow, o/p volts = are >less than the grid, then it would draw power from the grid. When the wind = picks >up, the turbine would then have to "pay back" power into the grid so as to >break even? > >I'm just having trouble visualising how the two systems mesh together. As = I see >it, the grid's frequency (50Hz) is rock steady 'cause the rotors in the >generators at the (coal-fired) power station weigh many tons and their >rotational inertia doesn't allow the line frequency to vary much at all. >Likewise the 3 phases are locked in because their voltage is induced by t= he >moving rotor. Hope I'm on the right track? > >The wind turbines, on the other hand, are situated in a cluster on the coa= st. >The turbulence created by the front row must affect the windflow for the o= nes >behind, I guess? There'd be minor differences in angular speeds for each >machine due to differences in mass, frictional resistance and so on. Also = the >position of the blades, hence the position of each turbine rotor wrt the f= ield >coils, would be random. So the phases & frequencies would not only vary hu= gely >from the grid but they'd also be different from turbine to turbine? Yet the >output of the entire turbine farm would have to be synched with the grid v= /v >phase and frequency at a total power output of 12MW. Just curious how they= do >it 'cause it sounds like a "non trivial" problem to me. > >If they went for DC -> AC switching, they'd suffer pretty high conversion >losses wouldn't they? Assuming it could be done at Mega-watt levels. >Debbie > > > = >____________________________________________________ = >On Yahoo!7 = >360=B0 new features: Blog polls, visitor stats custom themes and more! = >http://www.yahoo7.com.au/360 > = > -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist