So sorry to burst your bubble about rooftop wind.  Yet another idea that looks good until you actually do the math—then it becomes a counter-productive waste of time and money and energy.  See http://www.nytimes.com/2008/09/04/business/04wind.html?ei=5070 for a good overview.

Unsubsidized costs are

Rooftop wind is around $1.50 KWH – that is around $300/bbl - $400/bbl oil for diesel generation (O’ahu generation is heavy fuel oil, maybe half the price of diesel)

Residential tower turbines (well sited) $0.50 KWH (or maybe a little less with great sites) – that is around $120/bbl oil

Utility scale turbines (well sited) $0.10 KWH – cost effective now

(BTW, this connects back up with an earlier discussion about putting lots of little wind turbines up on utility poles.  Now we know that doesn’t work either.  Too bad—but let’s keep thinking outside the box.)

But all these costs are energy only and don’t included the costs of storage.  Location, Location, Location. Yes for sure for wind.  But Storage, storage, storage for sure too for any variable and intermittent renewable source of electricity.  AND THE PRICES ABOVE DO NOT INCLUDE STORAGE.

For little residential PV and wind projects, you can skip storage as long as you do them but not your neighbors.  They don’t scale up without storage.

Frequency storage determines how much variable renewable you can attach to your grid.  With no frequency storage you can only get some small percentage of total generation from V&I renewables.  Maybe 10%, maybe 15%, it depends on lots of factors, but it is a small number.  And the smaller your grid, the smaller the max percentage.  That is why on Maui they have to feather the wind turbine blades at night—they have to keep the percentage from wind below a threshold where the wind causes blackouts and the total nightly load is smaller so they can’t get as many MW from the wind farm.

Frequency storage costs are around $1M/MWH.  Hard to make a rule of thumb, but you probably need a ratio of 1x or 2X frequency storage MWH to wind capacity MW if you want to achieve high penetration on wind, e.g. numbers above 25%.  And we are fudging the definition of frequency storage to include just enough capacity storage to have time to notice the wind isn’t blowing anymore, then notice is doesn’t look like it is going to start blowing in the next few minutes, start up the fuel-based generation, and bring that online.

For comparison, frequency storage for fuel based generation is ZERO.  Fuel-based generation has a “cruise-control” throttle like your car.  Set it and forget it and the generator puts out the MW needed.  (The enlightened reader may ask at this time why we can’t use the cruise-control of the fuel based generation to compensate for the variable nature of the renewable generation.  To a small degree we can and do—that is exactly how we get the renewable penetration up to 10% or 15%.  But the fuel based generation can only do so much—there are actually physical limits based on the inertia of fuel flow, etc.  To continue the cruise control analogy, when your cruse control is set your car could go up and down slight hills and keep to the set speed.  But when you encounter a really steep hill the car will bog down before it can catch up to the right speed.  The utility grid can’t bog down from its 60HZ speed or your appliances (and other generation sources attached to the grid) blow up.

Even with infinite frequency storage, you can’t save the wind power for when the wind isn’t blowing.  So if the wind only blows 50% of the time even with an infinite wind farm and infinite frequency storage, you are only going to achieve 50% wind penetration.  If you want to save that blowing wind for later you’ve got to have capacity storage.  Batteries like we used for frequency storage don’t do it.  They are too expensive.  We need to use either compressed air storage or pumped water storage.  The prices are very hard to estimate and pumped water storage has a huge environmental foot print.)