In a posting “Accelerated Depreciation” Bishop hill says

This article at a blog called Billo The Wisp is important if true. Turbine gearbox failures apparently happen typically after 5-7 years rather than the 20 years that we are normally led to believe wind turbines last for. Moreover, their failure can be completely catastrophic, leading to the destruction of the whole turbine.

My comment is quite sceptical.

I do not think that the thrust of this post is correct – that there is a problem that gearboxes in that they will only last for 5-7 years, that has been around for 25 years and that it was so serious that the US government set up a special department to investigate in 2007. Despite all of this, there is still a largely hidden and hugely costly problem of which people are not aware. Having been in the engineering industry for a number of years I would consider the following if involved in the decision to set up a wind farm.

First, wind turbines are electro-mechanical devices. They need servicing and occasional overhauling. Ease of maintenance is important, including the replacement of major components. I would want a recommended maintenance program, along with projected parts costs, required maintenance equipment (e.g. a crane) and standard labour hours.

Second, I would want data on long-term historical performance, service and maintenance costs of each manufacturer’s equipment.

Third, if there was a large wind farm, I would include some spare parts, including major components that should last the life of the equipment. This may include have complete sets of spare parts that can be quickly swapped out – so major maintenance can be done in a workshop and not 200 metres in the air.

Fourth, I would cross-check this against industry journals. Wind turbine manufacture is a huge business with a number of manufacturers selling into a large number of countries. Issues are discussed, like in any industry.

The largest wind farms cost hundreds of millions. Businesses are not naïve. Even with large potential profits, there is always more money to be made through proper investment appraisal and protecting that investment through a thorough maintenance programme. If a major component of a wind turbine only lasted a third the length of time of the main structure, then replacing that component would become a part of the life-time costs. There would be huge incentives to minimize those costs through better design, such as ease of replacement of bearings. The only issue is that the real costs of wind turbines will never come down to a level where subsidies are no longer required.

NB a source of the reliability claims is this June 2010 article, which is now 3.5 years old.