Tim Worstall

Got a Calculator?

This plant is some 100 clicks away from us so we won’t be getting any of this power. But anyone care to work out how efficient it is? In terms of costs?

The world’s largest solar power plant has been inaugurated in Portugal. The 11-megawatt, £40 million plant, a joint project of GE Energy Financial Services and PowerLight Corporation of America, and the Portuguese renewable energy company Catavento, is located in Serpa, 200km (124 miles) southeast of Lisbon.

Southern Portugal has as much as 3,300 hours of sunlight a year and the plant will produce enough power to supply 8,000 homes. It will also prevent the emission of 30,000 tonnes of greenhouse gases a year when compared to fossil fuels.

In

10 responses

  1. Mark Brinkley Avatar
    Mark Brinkley

    An 11 megawatt PV array in Portugal should be capable of producing nearly 11mWh/annum, with a value around £750k at UK retail rates. Payback on £40m investment therefore around 50 years, pretty typical for PV installations. The anticipated lifespan of a PV arrays? Around 25 years.
    What we haven’t been told is how much grant money is included here.

    Reply
  2. Andrew Avatar
    Andrew

    MB – You mean 11 GW-h, I think (your numbers work that way).
    The 3300 hours of sunlight quoted in the Times story would suggest 48 GW-h per annum, though “sunlight” perhaps doesn’t mean the same as “solar plant running at maximum output”.
    Using your 1000 hours, and discounting, ignoring maintenance costs, the NPV of a 25yr life comes out at GBP12mm, rising to 21mm if energy prices rise by 5% per annum.
    OTOH, as I mentioned a while ago, the chief public benefit of this sort of project is, arguably, that it advances the state of the art, and makes the next one cheaper.

    Reply
  3. sanbikinoraion Avatar
    sanbikinoraion

    I did read something about them making an order-of-magnitude improvement in the efficiency of solar cells recently. Won’t be market-ready for another 5+ years ofc.

    Reply
  4. David Gillies Avatar
    David Gillies

    Order of magnitude? Nah. There’s no room for a tenfold increase. Double the efficiency, maybe. But even if we parse ‘order of magnitude improvement in efficiency’ to mean ‘order of magnitude decrease in inefficiency’, then that would imply going from, say, 15% efficient to 91.5%.
    Energy output over a year assuming the 11MW stated output is attained over the full 3300 hours of illumination = 11000 × 3300 kWh = 36 million kWh = 36 GWh. At 5% discount rate and ignoring operating expenses, break even for a £40 million asset over a 25 year lifespan requires revenue of ~£2.75 million p.a. (40 million divided by Sum[0.95^i, {i,0,24}]). This indicates a required price TO THE GRID (not the consumer) of ~7.6p/kWh. That’s higher than the rate to the consumer (5–7p).
    Note I have used very generous assumptions here. So, on the face of it, it would appear that this thing is a white elephant.

    Reply
  5. Josh Avatar
    Josh

    11MW? Oooh! Areva must be quaking in their boots. They’re only 2 orders of magnitude ahead with their feeble little 1.6GW PWR.

    Reply
  6. Mark Wadsworth Avatar
    Mark Wadsworth

    £40m divided by 8,000 homes = £5,000 per home.
    Amortised straight line over 25 years (taking this figure from Mark Brinkley above) = £200 per home per year. Looks good to me.

    Reply
  7. David Gillies Avatar
    David Gillies

    Aye, well, there’s the rub: 11 MW divided by 8000 homes = 1375 W per house, i.e. about enough to run a toaster (but not an electric kettle), so clearly nonsense. A household uses about 2.5 kW on average, but needs the ability to draw on maybe ten times this as a transient load.
    Solar is not, and never will be, baseload power supply. So talking about it powering x homes is fatuous. That 11 MW figure is the maximum power output. And 3300 hours is only 38% of the hours in a year, so two thirds of the time it’s not producing anything at all, even if we make the risible assumption that for the third of the year it is operating it’s generating the full 11 MW.
    Put it another way: 3300 hours generating 11 MW is 130 TJ, for a cost of joules per year per pound of 3.3 MJ/year pound (and remember this assumes 100% output for all 3300 hours). In a year, a 1 GW conventional power station will produce 10^9 × 8760 × 3600 = 31 PJ. Does anyone really think a 1 GW power station costs £10 billion, as it would have to have the same cost per unit of generating capacity?
    To put it yet another way, UK baseload (yes, I know this is in Portugal) needs about 40 1GW stations, or nearly ten thousand of these solar farms (if we could somehow arrange to a) get 3300 hours a year out of them and b) store the energy they produce during the day and dole it out at night).

    Reply
  8. Matthew Avatar
    Matthew

    If a household uses 2.5 kw on average, that’s 21,900 kwH per year, which at the 6p rate you say it costs to the consumer makes the average electricity bill about £1,200, doesn’t it?

    Reply
  9. David Gillies Avatar
    David Gillies

    Suunds about right to me.

    Reply
  10. Matthew Avatar
    Matthew

    Way too high, surely?

    Reply

Leave a Reply

Discover more from Tim Worstall

Subscribe now to keep reading and get access to the full archive.

Continue reading