Just adding to what other people have said with some bits from the article and my ramblings:
>energy losses in mechanical gears. These losses can be 1 to 2 percent per stage, and many turbines have three gear stages, explains Powell.
In order to have the generator at the bottom you would need at least 2 things, assuming your generator/gearbox is now going to stand vertically within the tower and take its drive directly in. 1) a differential to change the motion axis, positioned at the top. 2) a shaft that either hangs or is supported at intervals, for the length of the tower.
A couple of issues I can imagine with this,
The diff at the top needs to be of reasonable size. In mining and heavy industrial equipment 10MW gearboxes aren't unheard of, but the article also mentions that the turbine speed is quite slow, so the torque will be extremely high to result in that 10MW.
The 1-2 percent loss of this 'stage' will also necessitate a cooling system at the top separately just for this diff as 2% of 10MW is 200kw of waste energy (heat) that has to go somewhere.
However, an advantage is that you can gear the diff to spin the vertical shaft faster reducing the need for some of the primary gearbox gearing and reducing torque load of the shaft.
The shaft supports will also sap power as you have the weight of the shaft now on one or more bearings. The way I designed this mentally was having the shaft split in to sections, perhaps 5m each so 30 sections, as installing a continuous 150m shaft isn't realistic. Each shaft section with its own bearing for servicing and maintainability. This could be reduced to 1 bearing per 2-3 shafts based on losses vs ease of servicing a 15m section over a 5m section.
Other approaches that don't include a shaft could be considered, hydraulics, belts even.
>InnWind then had to add even more steel because of an unexpected resonance in the structure. This problem resulted from the mass at the top of the tower being so light that when the 41.7-metric-ton blades swung past the tower, they strained the structure at a frequency that was too close to its natural frequency.
Issue 2 raised in the article is that having the mass at the top offsets the weight and harmonics of the blade, moving all that junk to ground level may mean adding ballast weight to counter vibration.
>energy losses in mechanical gears. These losses can be 1 to 2 percent per stage, and many turbines have three gear stages, explains Powell.
In order to have the generator at the bottom you would need at least 2 things, assuming your generator/gearbox is now going to stand vertically within the tower and take its drive directly in. 1) a differential to change the motion axis, positioned at the top. 2) a shaft that either hangs or is supported at intervals, for the length of the tower.
A couple of issues I can imagine with this, The diff at the top needs to be of reasonable size. In mining and heavy industrial equipment 10MW gearboxes aren't unheard of, but the article also mentions that the turbine speed is quite slow, so the torque will be extremely high to result in that 10MW.
The 1-2 percent loss of this 'stage' will also necessitate a cooling system at the top separately just for this diff as 2% of 10MW is 200kw of waste energy (heat) that has to go somewhere.
However, an advantage is that you can gear the diff to spin the vertical shaft faster reducing the need for some of the primary gearbox gearing and reducing torque load of the shaft.
The shaft supports will also sap power as you have the weight of the shaft now on one or more bearings. The way I designed this mentally was having the shaft split in to sections, perhaps 5m each so 30 sections, as installing a continuous 150m shaft isn't realistic. Each shaft section with its own bearing for servicing and maintainability. This could be reduced to 1 bearing per 2-3 shafts based on losses vs ease of servicing a 15m section over a 5m section.
Other approaches that don't include a shaft could be considered, hydraulics, belts even.
>InnWind then had to add even more steel because of an unexpected resonance in the structure. This problem resulted from the mass at the top of the tower being so light that when the 41.7-metric-ton blades swung past the tower, they strained the structure at a frequency that was too close to its natural frequency.
Issue 2 raised in the article is that having the mass at the top offsets the weight and harmonics of the blade, moving all that junk to ground level may mean adding ballast weight to counter vibration.
/rambling thought experiment