There are about 250,000 wind turbines generating economically sound electricity world-wide today. But some anti-wind lobbyists claim that wind turbines consume 50% or more of the electricity that they produce.
The claim appears to originate with Eric Rosenbloom, President of National Wind Watch  and creator of the wind disinformation site aweo.org site . Among other claims, Mr. Rosenbloom says that wind turbines have massive heating bills to de-ice blades as well as lots of electricity consumption to start the blades spinning. Mr. Rosenbloom’s sources are an unnamed Swedish graduate student on a now defunct discussion forum and unlisted personal correspondence. Note that Mr. Rosenbloom’s credentials do not give any assurance of any technical or engineering background sufficient to make the claims he is making; he is a graphic designer and editor, running a small business in that field.  Other anti-wind advocacy sites reference Mr. Rosenbloom’s AWEO site uncritically and did not add any substantiation.
Max Rheese, Executive Director of two Australian organizations, the anti-wind Australian Environmental Foundation and the climate-denialist Australian Climate Science Coalition, likes to say that Australian wind turbines all have air-conditioning as his variant on the Eric Rosenbloom myth. He lists absolutely no sources for his claims, but Mr. Rheese is known for his whoppers around wind energy. 
Note that these claims are unsubstantiated by anything that could be called evidence. They do not reference checkable sources, they do not show their calculations, and their claims are not assessed by independent bodies.
So what’s the real story? Do wind turbines suck up vast amounts of electricity from the grid to start and cool or heat themselves? Or are they pumping out 99%+ net electricity with a trivial amount going to their operation?
Mr. Rosenbloom, Mr. Rheese and other anti-wind campaigners who repeat the myth of low net generation are at best overly credulous and at worst lying through their teeth. De-icing is just starting to penetrate the market, air-conditioning doesn’t exist and published and reviewed lifecycle cost analyses (LCA) following ISO methodologies suggest that electrical consumption by wind towers is insignificant, so the ratio is extremely high.
As always, breaking down the myth into its component parts is worth-while:
- De-icing systems
- Blade start
Let’s take each in turn.
De-icing systems are just now showing up on wind turbines
In some northern climes in the winter time, ice does build up on wind turbine blades. This can reduce aerodynamic efficiency of the blades by a few percent.
Mr. Rosenbloom, originator of the de-icing myth, made his claims between 2006 and 2009.
Let’s read a quote from a wind industry journal from February 2012. 
Among the highlights during the first day at Winterwind 2012 was a stage presentation of de-icing solutions by six of the world’s leading manufacturers. Five of these systems are brand new and some have yet to enter the commercial market.
- Two years ago this session would have been unthinkable, the market just wasn’t there, says Göran Ronsten, program coordinator of Winterwind 2012.
The total number of orders for de-icing wind turbines referenced was under 1,000 and the total number working at that time was under 100. When de-icing is operating in wind turbines, it will be operating at a significant net gain for generation in harsh winters.
So, there are almost no de-icing systems on wind turbines operating today, and there weren’t any at all when Mr. Rosenbloom, who continues to promote this, first made his claim a few years ago.
There isn’t any air-conditioning on wind farms in Australia
According to Epuron spokesman Martin Poole, their wind turbines use simple radiators like those in cars to draw heat away from blades when they are operating and don’t air condition them at all. Consultation with several industry representatives confirmed that there are no known turbine designs in Australia that feature air conditioning.
There are cooling products for wind turbines, but they aren’t commonly used. Australia isn’t known for being a cool place, yet major wind farms don’t use cooling products.
Given that Mr. Rheese provides no sources at all for his claim and given his colourful relationship with the truth on both wind generation technology and global warming, it’s likely fair to say he made it up.
ISO standard full-lifecycle assessments make it clear that blade start is irrelevant
Wind turbine manufacturers have a vested interest in accurate full lifecycle cost assessments (LCA) and publish these. To maintain ISO certification — a necessity for any manufacturer to sell to more than backyard hobbyists — LCAs are required to be published and available. They must follow manufacturing standard ISO 14040-14043 approaches to costing. They must be independently reviewed. What is relevant from these LCAs is that every step of manufacturing, transport, construction, operation, maintenance and dis-assembly is captured and quantified. Literally everything of any significance is measured and included in the calculations.
For the purposes of this, let’s look at one specific wind turbine LCA, for the Vesta 2.5MW onshore and offshore wind turbines.  As required, it has been created following ISO standards, published for anyone to read and reviewed by an accredited organization for accuracy.
As an example of the degree of rigour applied to this approach, here is the paragraph solely related to the cost of transport during maintenance and inspection activities:
It is estimated that inspection will be carried out four times a year per turbine. Three of these are assumed to be carried out by helicopter and the remaining one by boat. Inspection will also include about 2,400 km a year by car per turbine. Since no data for a helicopter is available, recalculated emission data from a mid-sized jet is used.
So what does this comprehensive, standards-based, reviewed LCA say about electrical consumption during operation? Nothing. The amount of electricity consumed during startup is so trivial compared to generation that it isn’t worth mentioning. It is insignificant.
Other LCAs for other wind turbines tell exactly the same story. Wind turbines consume so little electricity to start up compared to their generation that it just isn’t noted. It’s irrelevant.
Wind farm operators monitor this and net energy generation is 99.7%+
Hepburn Wind is a community wind farm in Australia. It’s Founding Chair, Simon Holmes a Court, has confirmed that over the first 11 months of operation, the wind farm generated 320 times as much energy as it consumed from the grid.
Outside of short duration draws at start up – and remember, wind turbines generate electricity 75-85% of the time and will often continue generating for weeks, so start up doesn’t occur that often – wind turbines generate all of the electricity that they consume.
How does this compare to other forms of generation?
All power stations draw power from the grid when they are not operating, and all draw power from their own generation when operating (known as parasitic load).
As a December 2013 Australian study says:
Electrical energy generated (usually shortened to ‘electricity’ or ‘energy generated’, or just ‘generation’) is the quantity of electricity produced in a given time period by generators, measured at the output terminal of each generating machine. Some of this output is used in the power stations, to drive, for example, pumps, fans, conveyors and crushers – collectively called auxiliaries. Auxiliary loads are typically up to nine per cent of electricity generated for a brown coal power station, five per cent for a black coal station, two to three per cent for a gas fired power station, and effectively zero for hydro and wind power stations.
This diagram from the study is instructive as well:
Along with the average 7% auxiliary power draw from generation plants — which is declining with each MW of wind energy and solar that is put on the grid –, the transmission loss of 2.7% is very instructive. Long range transmission of electricity is much more efficient than typically presumed. Distribution losses outweigh it greatly, but still the combined loss of transmission and distribution is only 7.4%. And this is in a large country, so distances are great.
Here is a useful summary diagram of the comparative results with other forms of power generation. 
Similar LCAs for GE support this analysis. , An interesting chart from this material is: