GE Grabs Gearless Wind Turbines

@ms
Thanks for the suggestion. That sentence has been fixed to make it a bit clearer.

The vertical wind turbines have so far had far lower efficiencies than the large tree bladed turbine designs that are the most common today.

The dramatic reduction in efficiency is a much bigger economical disadvantage than the advantage in very slightly lower maintenance cost by having the generator at ground level, and eliminating the minor cost of the yaw mechanism...

What's needed is an efficient design.

Artemis Wind is promoting a fluid drive for wind turbines, http://www.artemiswind.com/

A few years ago, Parker Hannifin received a grant from the Canadian government to develop a fluid power gearbox for wind turbines.  I haven't heard of any progress on it.

I believe the main disadvantage of all direct drive wind turbines, and most variable speed geared turbines with a fixed ratio gearbox, is that the generator speed must change directly with the rotor speed.  This produces "wild" AC, which must be converted by power electronics into constant frequency power for the grid, as you mentioned.  In other words, there is no way to eliminate the power electronics with direct drive or conventional geared variable speed wind turbines.

AAER and DeWind, possibly others, offer geared turbines with a variable speed gearbox, in order to drive the generator at constant speed even though the rotor speed is variable.  This eliminates the power electronics and allows the use of a synchronous generator, which is the kind used in central power stations.  Voltage and power factor are controlled by varying the field.  The synchronous generator is directly connected to the grid.  They are simpler, more efficient and, I understand, intrinsically more robust than the induction type that must be used with power electronics when the generator speed is variable, as it is in most wind turbines today, geared or direct drive.

About twenty years ago, in order to obtain constant frequency power with a fixed ratio gearbox, some designers achieved constant rotor speed by varying the load on the generator, as you suggested.  I believe the NASA turbine at Plum Brook on Lake Erie was of this design.  It was razed several years ago because it had long since been outmoded, and maybe because it was an eyesore with its two-bladed rotor, crude exterior and stumpy profile, although this is only speculation.  These turbines used synchronous generators for full control of the load and power characteristics.  In other words, they did not use power electronics, which was probably the main point of this simple arrangement.  In those days, and I am old enough to consider them recent history, power electronics were crude, expensive, inefficient and unreliable.  The problem with this old technique is that, compared to modern variable speed turbines which have the advantage of decent power electronics, it is not efficient.  To extract the most power from the wind, the rotor speed must change with the wind speed.  Therefore, power production in a turbine of the old design would have been considerably less than that of a modern variable speed turbine of similar size.  I doubt that any turbine of the old design is available today, although modern variable speed turbines probably control rotor speed to some extent by varying the generator load.  Mostly they use other techniques to match rotor speed to wind speed.

I doubt that direct drive is the best solution for wind turbines.  The generators are very large.  This makes them more difficult to seal against the weather, especially offshore.  A large generator also increases tower stress, hence cost, from the wind load - top hamper, to use a sailing term.  They are very heavy, and it doesn't seem likely that they can be made much lighter.  The greater the tower top weight, the more costly the tower, foundation and erection, although elimination of the generator might be enough to offset the additional weight of the generator.

However, if a standard synchronous generator is installed at the bottom of the tower, where it should be, driven by a vertical drive shaft connected to a variable speed gearbox at the top, then there is no need for power electronics.  The tower top weight, size and windage is minimum, hence cost is much reduced.  Maintenance is easier.  Therefore I predict that direct drive turbines will go the way of those constant speed wind turbines of twenty years ago, at least for turbines of more than 2 MW.