Windmills vs. Photovoltaics

We have attempted to estimate the contribution of photovoltaics to global energy production. Currently the installed base of photovoltaics worldwide has an output of about 5.0 gigawatts. Since the sun doesn’t shine 24 hours per day, the actual yield is probably about one-third that amount, call it 2.0 gigawatt-years in 2005. While this installed base is probably set to double every two years, and this amazing pace may be a conservative estimate, it is accurate to say photovoltaics do not currently contribute to global energy production in any meaningful way.

We are moving to an electricity future – and since 33 gigawatt-years equals one quadrillion BTUs, and since global energy production in 2005 was about 400 quadrillion BTUs, today we are only getting about one-6,000th, or 1.7 hundredths of one percent of our energy from photovoltaics.

What about wind power? According to the Global Wind Energy Council, the installed base of windmills in the world today have a cumulative output of 59.1 gigawatts. Since the wind doesn’t blow 24 hours per day, the actual yield, again, is probably a bit more than one-third that amount, call this 24 gigawatt-years in 2005.

So wind power is currently producing nearly 12 times more power than photovoltaics, which brings their share of global energy production up to a measurable but still paltry one-500th, or 20 hundredths of one percent.

Which of these eminently clean sources of energy are increasing faster? In 2005, 11.5 gigawatts of windmill generators were manufactured. In the same year, about 1.5 gigawatts of photovoltaics were manufactured. But the costs of windmills are not likely to drop in the future as much as photovoltaics. We predict that by 2010 global production of photovoltaics will outstrip that of windmills, and that will just be the start.

It is reasonable to assume wind power has the potential to increase output by an order of magnitude, to perhaps 1.0% of global energy production. This is based on the assumption that mega wind farms such as those just announced in the Thames Estuary, will continue to get built. Should decentralized, silent, ducted wind-generators be developed and widely adopted for use in commercial buildings and urban high-rises, perhaps wind energy could even reach 2.0% of total global energy production within a generation, perhaps much more. But there is an upper bound to how many places are truly suitable for windmills; there are problems with aesthetics; there are problems with birds; there aren’t windy regions everywhere; there are clean energy alternatives.

Because the price of photovoltaics continues to drop – thin film costs are already approaching the magic $1.00 per watt level – combined with lower maintenance costs compared to windmills, our money is on photovoltaics. There appears to be no limit to how much energy will eventually be produced by photovoltaics, and production of photovoltaics is poised to experience sustained exponential growth.

11 Responses to “Windmills vs. Photovoltaics”
  1. Michael says:

    Please stop using the term “windmills”. The perpetuates yours and your reader’s ignorance with regard to this technology.

    The commercial installed cost threshold of $1/watt is the key not just the manufacturing cost. This should be made clear.


  2. Ed Ring says:

    We could call them aerogenerators, I suppose. What is your preferred term?

    The $1.00 per watt threshold for photovoltaics was referring to manufacturing cost. Installation costs are another story. A lot of energy and innovation has gone into reducing manufacturing costs for photovoltaics but installation costs remain at $5.00+ per watt for most photovoltaics. Getting installation costs down is being addressed, especially as volumes increase, and as you say, that is a separate and equally relevant issue.

  3. Rosa says:

    Money cost aside, do we know how much energy it takes to manufacture a certain area of PV cells (by any of the current methods), and how much energy they produce in their lifetime (which is limited, mind you)? Using a GWh of mostly coal power to produce PV cells that yield a GWh worth of energy trickled over the course of 30 years doesn’t sound particularly green to me.

  4. Ed Ring says:

    That is a good question, and one we haven’t addressed in a while. Here’s a study “What is the Energy Payback for PV?” that attempts to answer this.

    In this study, which is consistent with others we’ve looked at, a PV system will return about 20x the energy that went into it during its lifetime output. This multiple, which is far better than anything reported to-date for biofuel, continues to improve as thin film technology becomes commercialized.

    Moreover, since the primary energy required to make PVs is electricity, and PVs produce electricity, it is possible from an energy standpoint to have photovoltaics themselves generate the power required for the expansion of their own manufacturing output.

  5. Jeremy says:

    One of the challenges in SOLAR VERSUS WIND is the comparison based on nameplates is just wrong. Glad you spotted that counting nameplate power numbers need to take into consideration whether it’s sunny or windy. But additionally, nameplate values don’t translate to power produced by simply dividing by 1/3rd.

    It’s the net power production UNDER THE CURVE over time that matters. Some cells will be much more efficient across the spectrum, angle of incidence etc. than others. Some turbines have better startup scenarios, better capture in light winds, etc. And I can state from authority that two nameplate wind turbines of 3kwh might produce 3200 or 6400 kwhrs per year (a 100% variance).

    So, someday, we’ll move from the useless “nameplate” value of noon-day clear skies Arizona solar and 35 MPH wind at 30 meter hub height, and get to YEARLY POWER PRODUCED, which is what we are buying these things for.

    It all reminds me of “page views” in the Internet web days. The irony being, Google – the most successful Internet game of them all, monetized the virtual opposite construct – click-throughs.

    One other comment: $/watt won’t be the only math, in urban environs – where an increasing majority of the world lives, watts/sq meter will also matter. Case in point, Google is building one of the largest deployments, campus wide, and it will generate a measly 1.6 mw (nameplate) – in a good windy campus, a wind system might be able to do 10-100x that power production on the campus size of google. (or more concretely, there are approximately 36 watts / sq meter on a rooftop in solar power, and 320 watts / sq meter on a rooftop in wind power (even higher for wind if you are not height constrained).

    Of course, I work for an urban scale wind power startup, so I don’t come unbiased into this game.

  6. Simon says:

    Just so you know…

    Wind Power is very much a reality…

    It covers app. 20 % of the Danish power consumption anually. Weather or not wind power or PV is the appropriate choice on a certain location depends on the sun intensity and hours and wind velocities.

    So the the choice isnĀ“t one or the other, but one in one location and the other in another location.

    This website is in Danish:,dk/

    It says that we cover 20 % of our consumption with wind power and in 2015 we should reach 35 %.

    This is simply a result of 10-15 years of policy to promote and further developing wind power production.


  7. Gary Moss says:

    We should be focusing on this more, not arcane babbling about reducing carbon dioxide output simply for the sake of reducing carbon dioxide output as so many blogs do, where people with a limited grasp of reality arcane points as a matter of status quo in a computer society of all things! Good story btw.

  8. Fred Irani says:

    How about the energy required to create the high tensile steels required for manufacture of the generators used by “wind mills” (or whatever)? Do we get as much energy out of a windmill as we put into its manufacture, installation and maintenence, -social costs aside?

  9. Andreas says:

    I like this discussion a lot but as a fellow Dane, I agree with Simon. The question is not one or the other, but both. We need to harness erenewable nergy from the globe in everyway we can – and each alternative has it’s own niche. Since there is not so much sun, but a lot of wind in Denmark, Denmark has become the leading country for wind turbines. In the alps, billions of watts are being pulled from damns and river turbines and in sunnier countries, PV’s could be preferred.

    By the way. Simons site is also in english:,en/

  10. Jon Bohmer says:

    Don’t forget solar thermal. The world also needs heat energy, and the output from solar water heaters are currently 30 times higher than PV. Plus, this industry grows at 40% per year.

    The PV numbers do not include solar thermal plants either.

    As to the growth of PV, an example is a new plant being constructed by Norwegian PV company REC in Singapore. Its output will be about 1.5 GW per year. If these PV cells are used in 10x solar concentrators the output is suddenly 15GW per year. The use of lenses and mirrors will lower the cost and increase the output of solar to become meaningful. In 10 years time, it is likely that we won’t see traditional un-concentrated flat plate PV anymore.

  11. Ben says:

    Well, if you think about the energy input into getting oil and coal, ask why people haven’t thought of this before.
    If anyone could tell me why coal is a good idea, show me. First they destroy the land, and then burn it and destroy the atmosphere.
    And don’t say because coal is readily available. because I can’t remember the last time I went looking for a sun burn!!!
    And how many times I voluntarily chased my hat after the wind blew it off!!!! I live in Ohio, and Ohio isn’t the windiest nor the sunniest place in the U.S.


Leave a Reply

You must be logged in to post a comment.