CO2 and backup plants

What I\’d be interested to know is whether this is in fact true?

Centrica and other energy companies last week told DECC that, if Britain is to spend £100 billion on building thousands of wind turbines, it will require the building of 17 new gas-fired power stations simply to provide back-up for all those times when the wind drops and the windmills produce even less power than usual.

We will thus be landed in the ludicrous position of having to spend an additional £10 billion on those 17 dedicated power stations, which will be kept running on \”spinning reserve\”, 24 hours a day, just to make up for the fundamental problem of wind turbines. This is that their power continually fluctuates anywhere between full capacity to zero (where it often stood last winter, when national electricity demand was at a peak). So unless back-up power is instantly available to match any shortfall, the lights will go out.

Two things make this even more absurd. One, as the energy companies pointed out to DECC, is that it will be amazingly costly and wildly uneconomical, since the dedicated power plants will often have to run at a low rate of efficiency, burning gas but not producing electricity. This will add billions more to our fuel bills for no practical purpose. The other absurdity, as recent detailed studies have confirmed, is that gas-fired power stations running on \”spinning reserve\” chuck out much more CO2 than when they are running at full efficiency – thus negating any savings in CO2 emissions supposedly achieved by the windmills themselves.

That we need to have backup gas stations, yes, agreed that that is obviously true. There\’s no other energy production (or, rather, energy conversion system into \’leccie) that is sufficiently responsive to be able to act as back up.

But is it actually true that the CO2 emissions from such back up are going to be greater than the emissions saved by having the windmills?

Yes, I know, there\’s a number of people saying it will be. And from the other side, well, I\’ve not really seen a discussion of this from the pro-windmill side.

So, can anyone point me to a reasoned and reasonable discussion of this point: what are the emissions from the required back up?

No, people telling us that a smart grid will do it all are not good enough. Yes, I know that Greenpeace et al start their calculations by saying we\’ll be using less energy in 40 years than we do now. And that energy will be intermittent at the discretion of the grid (that\’s what those smart meters allowing power to be cut to houses at times of peak demand are all about). That\’s not what I mean at all.

What are the emissions from gas fired back up as against emissions not made by having windmills?

20 thoughts on “CO2 and backup plants”

  1. It’s not true that we need the gas fired stations as spinning reserve (doesn’t that sound like a New Labour emergency team?) for when there is no wind.

    The governments preferred solution to the variability of wind power is demand management. If you’re sarcastically asking yourself “what are they going to do, turn out my lights if the wind doesn’t blow”, the answer is yes. Demand management is making the supply to users dependent on the wind blowing.

    Welcome to the middle ages.

  2. Mr Cameron, unwittingly provides the answer.

    It is a bit like Mr Cameron famously “reducing” his CO2 emissions by riding to Parliament on a bicycle, but with his car following behind, becasue it is still needed as back up.

    Notionally Mr Cameron is producing no CO2 emissions on his bike and but the car is. And even allowing that the car should be producing less CO2 emissions because Mr Cameron is not in it, it has to follow slowly, its engine is not working within its most efficient range, so
    there fore it in fact produces more emissions than if Mr Cameron had left his bike at home and gone to work in the car.

    The net result is Mr Cameron and his bike stunt produces increased CO2 emissions.

    Of course there would be an all round CO2 saving if Mr C used his bike with no car following.

    Similarly if wind turbines replace equivalent output from gas generators, there would be an overall saving in CO2 emissions, but in this case there would be periods when the grid would shut down becasue of the fluctuations in input from wind power.

    To prevent this, back up is needed: sufficient generators with capability to replace the total output from wind. These would be connected to the grid and supplying some load, part of their capability, ready to take over as needed within seconds.

    As with the car following Mr Cameron, notionally their emissions would be reduced proportional to the amount of energy being created by wind, except like the car they will not be operating in their most efficient range and will in effect be making the same amount of power anyway – the needed potential difference at 50Hz to match the load on the grid from wind – it just will not be used.

  3. John B,

    and of course Mr Cameron is also producing CO2 when riding his bike (although maybe less than a car)

  4. “gas-fired power stations running on “spinning reserve” chuck out much more CO2 than when they are running at full efficiency”

    I’m not any sort of expert, but that seems like it must be wrong. If it’s producing more CO2, it’s presumably burning more gas. And the point of “spinning reserve” is presumably to save money (by saving gas) – if it didn’t then surely the power company would run them at peak efficiency all the time?

    Remember this is a journalist’s interpretation of a PR hack’s claim about a scientific paper; there’s plenty of room for mistakes along the way.

    What would sound more possible would be if there was a CO2 saving from the difference between “spinning reserve” and “peak efficiency”, but that it was less than the CO2 cost of producing, installing & operating the wind turbine.

    Tim adds: No, this is getting to the heart of the assertions. Which is that, in order to have turbines which you can spin up in seconds you have to have turbines of a certain type, ones less efficient than those which you would ramp up more slowly.

    This extra inefficiency is such that it overwhelms the CO2 savings from having the windmills for the 25% of the time they do work.

    That, at heart, is the claim: I’d like to know whether it’s true.

  5. There’s a company (Wartsila?) promoting a new gas turbine which can be cycled on and off very quickly. So the spinning reserve problem may already be on its way to being fixed?

  6. I think we should use wind power to run Bitcoin mining rigs. I would gladly accept a subsidy from the state to do this and I would accept that the rigs have to be shut down when there is no wind available. It could be done with a smart grid.

    Nothing much to do with the original question, sorry. Power generation economics is well beyond my tiny mind.

  7. Can anyone help me here?
    Greenpeace and the govt assume that a wind turbine saves carbon as soon as it starts generating. This is nonsense of course because the inputs (rare earths for the magnets, steel, copper, concrete etc) need a lot of energy to mine, refine, transport and install.
    So using carbon input / carbon saving as our sole unit of account, what is the average time before and average wind turbine becomes net carbon saving?
    Every site I’ve found seems to want to include some complicating factor, but surely this isn’t Impossible Science and the answer can be found somewhere?

    Tim adds: The number I happen to know is the lifecycle emissions. Forgotten what the unit is but wind, hydro and nuclear are all 14-18 things per unit of whats. Solar about 35, gas 300 and coal 500.

    But that is leaving out the back up power for wind and solar.

  8. GE sells a gas power plant technology that can ramp up from “off” to “full power” in 10 minutes. Given that wind, sun and electricity demand forecasts are all accurate to within more than 10 minutes, this answers the question .

    The question of whether the carbon and/or capital embedded in building a turbine you only use for peak demand outweighs the carbon savings from using wind turbines is not answered by this analysis. But the question Tim asks at #5 above is conclusively answered “no”.

    (BTW, “John B” with no link, please can you use another name here? I’ve been commenting on this website and its predecessors under that name since 2004, so I don’t think I’m the one who oughta change…)

    Tim adds: No, we’re still not solved yet. We don’t know the relative efficiency of a turbine that can be ramped up that quickly as against one that isn’t ramped up and down. It wouldn’t surprise at all to find out that the ability to ramp up and down that quickly came at the expense of some general efficiency. Analagous to but not the same as cars which are great at accelerating and braking aren’t as efficient as ones optimised for general 60 mph efficiency working at 60 mph.

    And that is one of the things that is being asked: are the responsive turbines so much more inefficient than the unresponsive ones that this inefficiency more than covers the CO2 savings of the windmills 25% of the time?

  9. There are small gas turbine stations in Letchworth and Southall, and maybe some other places, which can be up to speed in about 30 seconds. The catch is they run on very expensive jet fuel, not natural gas, so they are only used when the grid controller humans are absolutely desperate for power. A quick look at shows that they have all closed down.

    There are also hydro pumped-storage stations in North Wales and Scotland which also have a fast response time, but there is no way you could use them as spinning reserve as far as I know unless someone knows differently.

    So we’re stuck with running large coal or combined-cycle natural gas stations for spinning reserve.

  10. This 2007 report which discusses the hidden fuel costs of German and Dutch wind generated electricity has some calculations and tables on how backup power efficiency negates any CO2 saving at the turbine. A relatively small installed base compared to conventional power will have little effect but as the installed renewable base becomes larger the conventional power has to respond to larger variations and becomes progressively more inefficient.

    Link to report here

    Of course Germany produced/s a lot of her power from coal so using 100% gas instead with their better ramp up/down capability and less CO2 per unit of electricity means that the inefficiencies do not have such a negative CO2 effect.

    Not mentioned here is the additional CO2 produced in the production and maintenance of wind turbines which is significantly greater that for equivalent conventional power station(think of equipment footprint and distribution size). Excluding the negative effect from backup power generation others have quantified an overall negative CO2 output for wind farms based on a whole life cycle.

    Bad news all round if you think CO2 is the be all and end all of warming temperatures. I however am looking forward to the new scare, that approaching maunder minimum and skating on the Thames again.

  11. Tim, discussion on this is limited because there’s actually very little data out there. The studies quoted by the anti-wind lobby – e.g. the one given above – tend to insert a random integer for the value of efficiency lost by frequent ramping actions. Shockingly this produces the result that wind power balanced by gas is more carbon-intensive than gas alone.

    Of course, if you insert another random number, you get a different answer. The most comprehensive literature review on this is from 2006, and can be found at: It indicates that inefficiencies caused by frequent ramping actions can reduce the carbon saved by displacing fossil fuel savings by as much as 7%. Gosh. Clearly that’s an issue.

  12. I’d say it is 50% guff. Gasfired power stations can be powered up very quickly unlike coal or nuclear which use great big boilers to drive turbines, and they don’t have to be on hot standby because of changes in the amount of wind. Demand varies much more than the amount of power available across the country from wind and both are reasonably predictable. On the other hand there is some truth in Centrica’s case because the variation between the amount of power available on the windiest days is considerably more than the amount available on the least windy days, but we have the same problem with solar and tidal power and we have similar issues with the availability/price of gas in the winter.

  13. Johnb, those GE turbines are at power from spinning, not from cold. So although they *may* reduce the CO2 a bit, they certainly don’t eliminate it. Note that GE seem to be under the impression that they can generate without much pre-spin, but that’s just PR type “spin”. Furthermore, combined cycle plants are the efficient ones, and that means even longer spin up.

  14. On the distinction between open-cycle turbine and combined-cycle turbines, the question as to which will be used to balance wind depends on the predictability of output. A four-hour window means you can use CCGT, a half-hour window means OCGT. National Grid anticipate that they’ll have meaningful accuracy for four-hour prediction by 2020, meaning that we’ll be using CCGT for wind shadowing.

  15. Tim
    You are missing the point – we NEED the CO2 to help keep up the agricultura yield to feed everyone. Not to mention the fact that a gas which forms 0.039% of the atmosphere doesn’t affect the natural changes in the climate.

    Not to mention the fact that Windmills have to have very solid concrete foundations – and making concrete exudes MASSES of CO2.

  16. Sorry this is late missed this first time round.

    Wind energy is constantly variable, which is no good for the national grid, so it needs to be balanced by conventional plant in “cycling” mode. It is this that causes the plant to release more CO2 than they would do normally. There is a paper like “Bentek: when more became less” that has some real world figures from Texas and somewhere else. The net co2 impact of wind plant is often marginal and sometimes negative. The Bentek abstract is freely available, and they will send you a copy of the full paper for free if you ask.

Leave a Reply

Your email address will not be published. Required fields are marked *