How much more electricity do we need to go to 100% electric vehicles?

As reported in Blowout week 146 the EU is drafting legislation to mandate the installation of electric vehicle charging stations in new homes while Germany and the Netherlands are considering legislation requiring that all cars and light vehicles sold after 2025 or 2030 must be 100% electric. None of this legislation has as yet been approved, but if it is how much extra electricity will be needed to power the millions of EVs involved, and how much will it cost? I’ve seen no numbers on this, so in this post I present some, starting with Germany, the Netherlands and the EU and adding a few more countries – and the world – as we go. Because of the uncertainties in the data and assumptions used the numbers should be considered as ball-park estimates only.

First Sources of data:

I used the following data to make estimates of how much additional electricity and electric generating capacity would be needed to power the EV fleets:

1. Number of cars and light vehicles currently registered

I obtained these data from a variety of sources too numerous to cite. In some cases the numbers are not entirely reliable.

2. Average distance driven per vehicle per year.

As per 1.

3. Average consumption in kWh/100km

I used the median value of 22kWh (equivalent to 153 miles per US gallon according to the US EPA) obtained from 114 test values published by the US DOE.

4. Average capacity factor of the additional electric capacity needed.

The idea of going to EVs is to reduce GHG emissions, so I assumed that the additional capacity would be “new renewables” – dominantly offshore wind and solar PV. After checking various published numbers I estimated an overall capacity factor of 30%.

5. Average cost of additional capacity in $/kW installed

After once more checking a number of capital cost estimates I estimated an overall cost of $3,000 per installed kilowatt.

I could, however, make no estimates for the following items, which could – almost certainly will – significantly increase costs:

6. The grid upgrades, domestic wiring upgrades, charging stations, smart meters etc. needed to distribute the additional electricity and charge/discharge the EVs.

7. EV purchase costs (which would probably have to be subsidized)

The idea underlying the EU legislation is that the EVs will act as storage batteries that can be charged from the grid during periods of low demand and/or high generation and discharged back into the grid during periods of high demand and/or low generation, thereby smoothing out the load curve while making full use of intermittent renewables generation. The estimates assume that this can be done with 100% efficiency, although it certainly won’t turn out that way in practice.

Finally, all costs are given in US dollars.

Now to the results:


Legislation requiring that only EVs will be sold in Germany after 2030 has been passed by the Bundesrat, Germany’s lower house, but needs to pass the Bundestag, the upper house, before it becomes law. If it is passed we get the results shown in the table below. Replacing all of Germany’s 44,403,124 fossil-fuel-fired cars with EVs would require a 31% increase in Germany’s electricity generation and a 40% increase in Germany’s installed capacity. The cost of installing this extra capacity would be $232 billion:

Note: 2015 total electricity generation is from the 2016 BP Statistical Review

But if Germany shuts down its coal and nuclear plants the situation becomes a lot worse. To replace the generation from this 60GW of lost baseload capacity Germany would have to install another 140GW of renewables, and adding this to the 77GW of capacity needed to service EVs increases the installation cost from $232 to $650 billion.


The Netherlands lower house has passed a bill requiring that all vehicles registered in the country after 2025 – five years before Germany – must be EVs, although the legislation won’t pass into law until approved by the Dutch senate, which is presently mulling it over. As shown in the table below replacing the Netherlands’ 8 million fossil-fuel-fired cars with EVs would require a 21% increase in electricity generation and a 24% increase in installed capacity. The cost of installing this extra capacity would be $27 billion. Costs are proportionately lower than in Germany because the Dutch drive less:


Norway was considering EV legislation earlier this year but has decided that market forces alone will achieve the desired result. (Assisted by government subsidies that covered about half of the total purchase price, a third of the vehicles sold in Norway in 2015 were EVs. The government is in fact now considering rolling the subsidies back.) As shown in the table replacing Norway’s 2.5 million million fossil-fuel-fired cars with EVs would require only a 7% increase in electricity generation and a 12% increase in installed capacity. The cost of installing this extra capacity would be $11 billion. Norway gets off lightly in relative terms because most of its electricity goes to industrial installations such as smelters and metal refineries.

European Union:

The legislation currently being prepared by the EU shows that it is thinking along the same lines as Germany and the Netherlands. If the EU eventually adopts a 100% EV policy the requirements will be as shown below. Replacing the EU’s 250 million fossil-fuel-fired cars with EVs would require a 34% increase in electricity generation and a 43% increase in installed capacity. The overall cost of installing this extra capacity would be $1.3 trillion. Requirements would, however, vary significantly from country to country.

Now to the countries and regions that are not actively considering 100% EV legislation. What would it take for them to go to 100% EVs within the next few decades?

The UK: A 36% increase in generation and a 49% increase in installed capacity, costing $140 billion.

The USA: A 29% increase in generation and a 44% increase in installed capacity, costing $1.4 trillion.

China: An 11% increase in generation and a 16% increase in installed capacity, costing $735 billion. Note, however, that these numbers increase to 19%, 27%, and $1.2 trillion if we assume that the number of vehicles in China continues to grow at 4.4%/year through 2030.

The World: An 18% increase in generation and a 30% increase in installed capacity, costing $5.0 trillion. These numbers increase to 26%, 44%, and $7.3 trillion if we assume that the number of vehicles in the world continues to grow at 2.7%/year through 2030.


To paraphrase Everett Dirksen, “a trillion here, a trillion there, and pretty soon you’re talking real money.” But none of the costs listed above, which will be spread over several decades, are beyond the financial capacity of the countries involved. The question is whether 100% EVs will solve the problem of renewables intermittency?

The Guardian has no doubt that it will, “the EU initiative …. would open the door to a futuristic world in which cars supply energy to Europe’s power network at all times of the day and night, balancing shortfalls from intermittent renewable energies when the sun is not shining and the wind not blowing.” And on the face of it there is indeed a lot to be said for having EVs do double duty as transportation and as a source of energy storage. (If the UK’s 25.8 million cars were all 30kWh EVs they would store 774GWh when fully charged, almost a hundred times as much as the Dinorwig pumped hydro plant.) In practice, however, there are the following problems:

•  To balance the “shortfalls from intermittent renewable energies at all times of the day and night” the EVs would have to be plugged in to the grid at all times of the day and night, meaning that they would never go anywhere.

•  Although it sounds like a lot, 744GWh is in fact enough to power the UK for only about a day. So after a couple of windless and sunless days a renewables-dependent UK would have 25.8 million EVs with dead batteries.

•  EV owners will be happy to plug their vehicles in when there is enough surplus generation to charge them but may conveniently forget to do so when the current flow is going the other way.

•  The weather being unpredictable, there is no way anyone can be sure that their EV will be charged up when they need it to be.

So if EVs are to double duty as transportation and storage a compromise must be reached that allows them to do both. The result will be that only a small fraction of the stored EV energy would be available for use at any one time, meaning that even if the UK had a 100% electric vehicle fleet it would still not solve the intermittent renewables problem.

There’s also the problem of the ancillary equipment needed to make the system work. Grid upgrades to handle the increased power flows will cost money, and installing charge/discharge plugs in millions of homes and apartment blocks won’t be cheap either. Neither will the subsidies that will almost certainly be necessary to induce people to buy EV s in the first place ($10,000/EV for 25.8 million EVs, for example, works out to $258 billion). And somehow all these millions of installations will have to be linked together so that charging occurs when surplus power is available and discharging occurs when is isn’t – provided that the vehicle is plugged in and not out driving around somewhere and that the owner has remembered to plug it in.

And also provided that bursts of non-synchronous energy from millions of EV batteries don’t crash the National Grid.

And further provided, of course, that everyone has an EV to begin with. Will the auto industry be able to manufacture and market enough of them? According to the NASDAQ projections shown in the Figure below there’s a good chance it won’t. NASDAQ projects that there will be only about 400 million EVs in the world by 2040. Somewhere around 2 billion EVs would be needed for a 100% global EV fleet in 2040:

The legislators in the European Commission, the German Bundesrat and the Dutch lower house are clearly laboring under the misconception that Europe is firmly on course for a bright green renewable energy future and that EVs will be a key part of it. They seem to be unaware of the subsidy rollbacks, policy changes and decreasing “clean energy” investment that will likely prevent this from happening. They are in fact following in the footsteps of California, which in 1990 passed legislation requiring that 2 percent of the vehicles sold in the state by 1998, 5 percent of the vehicles sold by 2001 and 10 percent of the vehicles sold by 2003 be zero-emission vehicles. The legislation had to be rescinded when the ZEVs failed to materialize.

This entry was posted in Energy and tagged , , , . Bookmark the permalink.

152 Responses to How much more electricity do we need to go to 100% electric vehicles?

  1. tobyglyn says:

    It’s odd that the Samsung Galaxy Note 7 battery issue has been called by some an environmental disaster due to the problems recycling the problem phones and batteries.

    All those new EVs will require batteries to be created and then (hopefully) recycled as they eventually wear out. No problem?

  2. pyrrhus says:

    Nice analysis of a completely absurd question. But what about the cost of repairing these extremely fragile and very expensive vehicles? Is there even conceivably the quantity of renewable energy available–I say no. Could enough lithium batteries even be made, with the shortage of lithium? etc…

    • singletonengineer says:

      Yes, Pyrrhus, this is indeed a nice analysis.

      I attempted, some months back, to derive a ball-park estimate along these lines, but ended up with a muddled head and far too many variables and not enough data.

      Roger has displayed again a remarkable ability to clarify the complex while maintaining contact with reality.

    • John F. Hultquist says:

      I’ve looked at several articles regarding Lithium and it is difficult for me to make sense of most of them. An older one (2010) claims:
      For realistic strategic planning purposes automobile manufacturers should model the material requirement at 2 kg to 3 kg of technical grade Lithium Carbonate per nominal kWh of PHEV battery capacity.
      Current global LCE production of circa. 100,000 tonnes, if available, would therefore be sufficient for 2 to 3 million PHEV batteries of 16 kWh capacity (GM Volt class).

      USA sales of new vehicles is about 17 million per year. How long does the battery last? Is it close to 5 or close to 10 years? Let’s say 10. So after 10 years there will be a need for 17 million replacements and 17 million new ones, and in the following years the same. This just in the USA.
      Refine the numbers and calculations any way possible. It is hard to see this working out.

      Also (March 2015) interesting reading:

      • Beamspot says:

        Hi, I wrote that in the Blowout coments, worth repeating that here:

        According to USGS, there are 14MMT of lithium (IIRC, in metallic form, about 5 times more in carbonate form).

        That amounts to approx 65TWh, or like 1000M Tesla S65, more or less the same amount of cars currently running on the world. At a mean of 15TW of world’s power consumption, that stands for less than 5 hours of world’s energy consumption.

        But according to USGS and Meridien International, the 7.5MMT from Atacama (>50%) reported by USGS, are fake, an scam, overstated by about 4.5MMT (by Chilean government).

        And according to Meridien, from the remaining 3MMT at Salar de Atacama in Chile, about half MMT had been already extracted.

        And lithium is not recycled never, so it ends totally lost.

        According to some sources I have in Salar del Hombre Muerto (Argentina, in the lithium triangle), it seems that the yeld to obtain lithium carbonate battery grade is about 70%. But there are also other consumers, like glass manufacturers that use lithium. That reduces the amount of batteries we would obtain.

        If we move to LiFePO, then we will need 20% more lithium (due the fact that each electron delivers 20% less energy). For LiS, the situation is even worse.

        Lead runs short much faster than Lithium.

        Rare earths are another issues. Motors can be manufactures without them. Async squirrel AC async, or external excitation AC synchronous don’t use them, but they are less efficient per KW. Permanent magnet AC Sync motors use lots of neodymium.

        Prius uses 1Kg, 60€ of Nd, and 120grams or 144€ of dysprosium. The bottleneck is dysprosium, that is also used in high power wind turbines, and its production is really limited.

        Dysprosium, as well as Nd, but at a lesser extent, generates lots of toxic waste, and many Kg of radioactive waste per Kg of them. Also lots of other RE elements that perhaps are not used, like Europium, that is one of the key ingredients of white LED’s, and the reason why those LED’s are cheap: dysprosium pays the bill of europium extraction and purification (it’s a byproduct).

        In short, the issues with reserves are really limiting, not only for EV’s, but also for RE.


        • Stuart says:

          The ultimate factor behind the rise of the electric car will be the scalability and the cost curve of EV’s.

          There are lots of interesting things going on with renewables and EV but ultimately it all boils down to the price/cost of reusable chemical storage of potential energy (rechargeable batteries).

          But the cost curve of reusable chemical storage is completely unknown at the scale necessary to replace liquid fuels.

          A good example of cost curves is solar panels. It is well documented that the cost of solar panels has fallen dramatically and continues to fall. This is often extrapolated to suggest that in the not too distant future solar power will be so cheap nothing will be able to compete with solar. The cost curve over time is L shaped (exponential decay).

          Except in the real world, this doesn’t happen. When something is being scaled up to a global size it will inevitably run into supply constraints. Solar panel production will eventually find a volume where the base commodities are relatively scarce. At this point the cost starts to rise again.

          All cost curves are in fact U shaped over time.

          An excellent example of this is oil. in the 1800’s technology drove the price of oil lower and lower until it became ubiquitous. Until one day it started getting harder to find the stuff and the price began to rise again.

          The same thing is going to happen with the base commodities that make up rechargeable batteries Because some of these materials are not particularly abundant.

          Elon Musk himself has said that he would need 100% of the world’s annual lithium output to manufacture 500,000 EV’s each year. But there are some 80 million cars sold around the world each year.

          So currently the if we diverted the entire global lithium supply to EV’s it would allow 0.6% market share of new vehicles to be EV’s.

          The mining industry is the elephant in the room when it comes to the EV revolution. Are governments out prospecting the entire planet and endorsing licenses for big open air mines so that we can realise this EV dream? No they are not.

          If you look at the commodities the world economy consumes, over 60% by weight of all base commodities are fossil fuels (coal, oil, gas). the world economy consumes more tonnes of gas than all metals combines.

          Think about that. The weight of natural gas we consume is heavier than all the metals we consume.

          To switch to EV’s we are going to dramatically increase the volume of metals we mine in order to reduce the fossil fuels we burn.

          Long before EV’s reach 40 or 50% market share battery production is going to be commodity constrained. The EV cost curve is not L shaped it is U shaped. Just like everything else.

          The only reason the world economy has grown so huge is because fossil fuels had such a wide U.

          There is absolutely nothing to suggest that the materials for any battery chemistry can be mined in the quantities required to build 10’s of million EV’s each year.

          Just how long does it take to prospect, license, sanction and dig out these mines that will be needed?

          The EV revolution requires battery production to increase by two order of magnitude (100 fold). Economists struggle to accurately extrapolate changes that are larger than 5%.

          The whole thing is just wonderfully naive.

        • ristvan says:

          Lithium may or may not be a limiting factor. What is certain is that after Chile, its cost goes up significantly. Rare earths are not a factor, because the alternative is switched reluctance electric mortors such as in all the Dyson products. Only ordinary steels and copper are needed.

  3. Dave Rutledge says:

    Hi Roger,

    A good exercise.

    I would prefer to call the Norwegian encouraging of EVs waiving taxes rather than a subsidy.

    I would have approached the calculation differently. According to the 2015 IEA World Energy Outlook, the power sector demand in 2015 was 5.1Gtoe, and the transportation sector demand was 2.5Gtoe. Accordingly the grid generation would need to grow by 50%. I would suspect that your calculation is biased because most current EVs are light vehicles, with few SUVs and no trucks.

    EVs have been much more expensive than the equivalent gasoline or diesel cars, so actually purchasing the EVs would be a large additional cost.


    • Hi Dave:

      As far as I can determine all the pending legislation applies to light passenger vehicles only. There are apparently technical problems with battery-powered heavy trucks.

      And I’ve not allowed for aviation, shipping or railroads.

      A number of people have objected to my calling Norway’s financial encouragement a “subsidy”, so I guess I’m going to have to admit defeat on that one. 🙁

      • singletonengineer says:

        If it looks, feels and smells like a subsidy, then it is a subsidy.

        Norway subsidises purchase of EV’s via tax reductions. Still a subsidy.

      • Hugh Sharman says:

        Norway encourages EVs with access to bus lanes, free parking and tax breaks. These “fringe benefits” are financially very attractive, until, of course, the bus lanes and parking seize up, as they must!

    • Wookey says:

      “EVs have been much more expensive than the equivalent gasoline or diesel cars, so actually purchasing the EVs would be a large additional cost.”

      This is true. And the extra cost is entirely down to the cost of batteries. Those are falling quite fast, but are also somewhat shrouded in mystery. Enormous battery-building capacity is being built around the world at the moment, which is part of the reason costs are falling.

      However EVs are much cheaper to run than ICE cars, so even at current battery costs and oil prices, over the lifetime of usage they can be significantly cheaper than ICE cars. That is why people will buy (are buying) them. Even without current subsidies some EVs are already cheaper. Once that line is convincingly crossed it seems likely that they will become standard quite quickly. (I saw a very useful comparison of US BEVs, PHEVs and ICE vehicles, with and without subsidy, over first 100,000 miles a couple of weeks back, but can’t find the link now. Sorry)

      Obviously predicting future costs (of petrol, diesel, batteries, and other fuels) is a tricky game, but given current trends an EV-dominated future seems likely. Also, given the need for decarbonisation and cleaner urban air the pure ICE is going to be legislated away anyway, at least in cities, whether alternatives are cheaper or not.

    • jfon says:

      Dave, you can’t make a direct comparison like that, because an electric motor is about 90% efficient, versus an internal combustion engine’ s twenty percent. If it loses as much power again in transmission and storage, it’s still four times better .
      As for trucks, greatly expanded use of rail freight would save a lot of fuel compared to long haul trucks, even if the locomotives weren’t switched to electric. For local delivery, there are some electric pioneers.
      Instead of looking at the renewables aspect of it, what would widespread electrification do for a grid with all-nuclear baseload ? Solarholics are always saying that nuclear is too inflexible to match with varying demand and intermittent renewables. Since renewables output varies much more than demand does, how much could steady baseload cover if you junked the ‘ unreliables ‘ and plugged the nighttime demand gap with battery charging ?

      • Anders says:

        jfon, you say David can’t make a direct comparison and then you go on to make one yourself that is just as misleading. An automotive diesel engine has an efficiency of 30-35%, that is in converting the chemical energy in the fuel to usable mechanical energy on the shaft. The electric motor has a 90% efficiency converting electrical energy to usable mechanical energy on the shaft. However, the electrical energy has to be produced by some other means. If it is from a hydrocarbon source you need to factor in that efficiency as well, and include losses in charging and discharging. All in all you are getting very close to the diesel engine efficiency when comparing the conversion from chemical to mechanical energy.
        Since we are talking about replacement of all vehicles the real power source will continue to be hydrocarbons, or at least a substantial part of it will be.

  4. Davis Swan says:

    As always you are the troll under the bridge bringing up inconvenient FACTS when all we want to hear about are the glorious and fantastical VISIONS of a pollution and CO2 free future. You act as if there were cold, calm, gloomy days in the winter in the UK. As if! Shame on you.

    • Willem Post says:


      What about the air conditioning load in Florida and southern California sucking on all those EV batteries, while people are at work, and no way to drive home at 5 pm.

      For insurance, one could take a trailer to work, loaded with battery system, not connected to the sucking system, and drive home on the battery power of the trailer.

      • Davis Swan says:

        It is truly remarkable to me that publications like the Guardian would be stupid enough (apologies for being frank) to conclude that cars that people need to drive home could be used as a source of electricity. I occasionally drive my e-bike 32 km to work and believe me it needs to charge all day for me to have any hope of getting home on it. Not to mention Euan’s point about every electrical outlet having to support two-way electricity flow and a complete rebuild of the electricity distribution system to manage all of that. So much unhelpful Greentech nonsense. You might find my recent post of interest.

        P.S. Hey Euan, I have added a direct Energy Matters link to my sidebar. Someday you and I and Paul Bach will be recognized for our great service to humanity – the Queen will commission bobble-heads of all of us with proceeds going to the worldwide grid stabilization fund!

      • Thinkstoomuch says:

        Another timely post.

        Willem for CA you forget that all that a/c load is going to be carried by solar. Well that seems to be the hope anyway. One of those things I will believe when I see it.

        It could happen.


      • Alex says:

        And heating, which is a bigger issue in most of the developed world. Hopefully EVs use heat pumps for heating, rather than resistance heaters.

        • Willem Post says:


          EVs using heat pumps?

          Would that be on Mars or the Moon?

          • Alex says:

            Every car with air conditioning uses a heat pump. I don’t think they’d work on Mars or the Moon, but are quite effective on Earth.

            The Nissan salesman told me the LEAF uses the air conditioning unit for heating. I’m not sure if that is correct, but it’s the logical thing to do. Resistance heating should be reserved for the seats.

          • gweberbv says:


            as far as I know newer version of the Nissan Leaf are equipped with a heat pump for cabin heating.

          • OpenSourceElectricity says:

            Yes, heat pumps are standard for heating electric cars.
            What is missing in the article are the savings. The investment in EV+renewable generation would lead to a drastic drop of spendings on imported fuels, wotrth also several billions per year.
            Also 30kWh is a tiny battery compaered to those promised for cars to be sold in 2-3 years. So 30kWh available for balancing additional to other dynamic loads like heating (+heat storages (, building mass, hot water) ) + ballancing effects of large grids sum up.

        • Willem Post says:

          Alex and gweber,

          You are right about heat pumps for air conditioning on newer models.

          Some of the cooling and heating is used for the cabin and some to maintain the battery at near constant temperature for optimum performance and longer life.

          Older EVs did not have battery heating and cooling systems, and performed poorly in hot and cold climates.

          That is a design detail of an EV.

          Roger’s article is about the exchange of LDVs for EVs

  5. Willem Post says:


    The below article deals with the real world of batteries.

    Charging/discharging losses are at least 15%, plus about an equal quantity of vampire losses.

    Travel………………………….20000……….20000………..miles/y assumed
    Energy from battery………..0.29………….0.29………….kWh/mile DC
    Energy charged……………..5800…………5800…………kWh/y DC
    Vampire loss……………………847…………..847…………kWh/y AC @ 8 mile/d
    Charging loss…………………..997…………..997…………kWh/y AC @ 15%
    Total through meter…………7644…………7644…………kWh/y AC

  6. Michael Kirby says:

    A nice exercise. I wonder if the numbers are slightly more favorable than I would expect. If we assume 1.4 Trillion dollars of new renewable electricity production to provide enough electricity to handle 280 million cars, that’s about 5000 dollars in capital expense per car.

    I don’t think the capability is beyond the industrial capacity of any of the states in question. It’s really a question of allocating resources there. either through tax policy, or consumer preference.

    It generally takes 20 years to roll over an auto fleet. So presumably this would be about 70 billion a year in capital expenses in the united states. This also feels a bit low.


    • Willem Post says:


      US NEW light duty vehicle sales are about 17 million per year @ $35000 each is $60 billion/y.
      If they were EVs, that would be about $80 – $90 billion.

      Average travel is about 12000 miles/year @ 0.30 kWh/mile, or 3600 kWh/year.

      17 million x 3600 = 61.2 x 10^9 kWh would be the electricity consumption increase/year, plus 7% for T&D losses, to get energy fed to the grid, plus 5% self-use to get energy produced.

      How many wind turbines and solar panels would that be? And what would be the cost?

      • Michael Kirby says:

        I think that is the analysis that was done here.

        I was just asking whether it was correct.

        Based on the analysis in the original article it seemed like it was 5k per car to add electric generation capacity.

        I thought that was low.


    • Don’t forget the grid upgrades, smart meters etc. They won’t put another zero on the end of your $5000 but they might put a one in front.

      • Willem Post says:


        NEW LDVs (cars, minivans, SUVs, 1/4 ton pick ups) would be about $800 – $900 billion EACH YEAR, as EVs, instead of $600 billion for traditional LDVs.

        NOTE: Up to 50% of these EVs likely would be imports.

        Assume nuclear, because by that time even the starry-eyed greenies in California will finally see the light.

        10,000 MW x 8766 x 0.90 (US CF) = 78,840,000 MWh/y
        Less Self-Use 5% …………………………74,898,000
        Less T&D, 7%……………………………….69,655,140; to meters.

        Cap cost $50 billion/y for NEW nuclear reactor capacity to be ADDED EACH YEAR, FOR EACH of at least 20 years, to accommodate changing out the traditional LDVs.

        Plus replacement and new-built chargers at about 17 million x 2000 = $34 billion EACH YEAR, FOR EACH of at least 20 years, to accommodate changing out the traditional LDVs.

        Plus grid expansion (nuclear-plant end + distribution end), say $20 billion EACH YEAR, FOR EACH of at least 20 years, to accommodate changing out the traditional LDVs.

        If wind and solar would be insisted on, that would be at least 2 – 3 times the nuclear and grid cap costs, PLUS that energy would be at least 2 – 3 times current wholesale prices, especially with much offshore wind.

        I am so glad, I will not live to see all that nirvana being enjoyed by TEN billion people.

        • Willem Post says:



          Some people say they want to get rid of those evil fossil fuels. That is easier said than done. Here is an example:

          – Tar is a waste product of refineries. Add aggregate and you get asphalt. No tar, no asphalt.

          – Would we go back to concrete roads, as in the 1940s and 1950s? Making concrete is energy intensive, and making it with wind and solar energy would make it very expensive.

          – Concrete embedded energy: about 1.5 million joules/kg, or 1.425 million Btu/metric ton, or 0.234 barrel of crude oil/metric ton, or 0.424 barrel of crude oil/cubic yard.

          – Would we have very small roads, or go back to dirt roads all over the US, Europe, etc.?

          The WORLD is in no hurry “fighting global warming”. World energy consumption from fossil fuels was:

          78.2% in 2011 of TOTAL energy consumption
          78.4% in 2012
          78.3% in 2013
          78.3% in 2014

  7. indeed, a nice and and simple summary, demonstrating again
    that oil is not replaceable for individual car mobility. While still leaving out
    all the other difficulties of car and battery lifetime and so on.

    Dave Rutledge proposed an even simpler way to estimate the impossibility:

    “I would have approached the calculation differently. According to the 2015 IEA World Energy Outlook, the power sector demand in 2015 was 5.1Gtoe, and the transportation sector demand was 2.5Gtoe. Accordingly the grid generation would need to grow by 50%.”

    Perhaps, what is missing .. how many new nuclear power plants or coal etc would be needed
    (especially during winter times.)

    anyway, critical minds should wonder now why those after all trivial facts have not
    entered into the political debate? How is it possible to find “scientists” being silenced to express
    such simple truth? Who holds the strings behind this e-car based illusions?
    It might be interesting to add the e-bike replacement option as an alternative, for the small fraction of people not capable to use more “healthy” normal bikes and walking.
    (according to advertisements muscle assisted e-bikes use 0.4kWh/100km).

    This aside, and in short, once oil gets rare, this way of individual car base freedom life for many people in richer countries will come to an end.

    What will those people do, when they realise that all the alternative “freedoms” have been
    taken already?

  8. Hans Erren says:

    Thank you for the numbers for the Netherlands

  9. Roger wrote ‘Now to the countries and regions that are not actively considering 100% EV legislation. What would it take for them to go to 100% EVs within the next few decades?’

    I have nothing against electric cars, per se. In fact, when I drove one, I liked it. But, If, eventually, they do somehow meet a large part of our personal transport requirements, it might well be for urban air quality reasons, driven by public policy.

    Quite properly, with the ‘Energy Matters’ audience in mind, Roger has looked at this issue from an angle that comes easily to an energy expert – electricity capacity and supply. There is another way of looking at it, though. We could just ask ourselves ‘What is it that would induce me, or, my friends, family, colleagues and acquaintances to buy an electric car?’ That question cannot be answered without opening up a large number of other questions relating to – among many other things – cost (and costs already incurred), convenience and alternatives.

    • singletonengineer says:

      Or, as the article says right at the top, a change to the law. That’s what is being proposed in Germany, etc, throughout Europe and, for the starry-eyed, globally.

      If manufacturers are forced to stop making ICE light vehicles, then where are you and others going to be able to buy one?

      Remember also, that this isn’t being driven by “energy experts” or “friends, family, etc”.

      It is being driven by well-meaning folk who have swallowed the Kool-aid about the future for batteries.

      The message in this article is not that we will all be buying EV’s in 10 years or that ICE’s will be gone for ever in 30 – it points out why this cannot happen. The answers to the challenges poses by climate change, if there are answers, lie elsewhere.

      EV’s simply can’t achieve what is promised on their behalf.

      Nor can wind turbines + solar, alone, power the world.

      And no, I don’t have the answers either. I’m a Climate Change Doomer.

      • Leo Smith says:

        I read that as :”The answers to the challenges posed by climate change, if there are answers? Lie elsewhere!

        Of course climate change poses no challenges anyway, and certainly not ones we can do anything but react to

  10. Leo Smith says:

    Readers might be interested in a bit of ‘armchair engineering’ I indulged in some time ago, trying to see what a post fossil post industrial society might look like and how it might work.

    The basic findings were that ion energy density, only one specific and highly impractical battery – lithium air – could in any way compete with hydrocarbon fuels for serious off grid mobile power, and I thought that synthetic hydrocarbon fuels made with cheap nuclear electricity would be a more likely way to keep Green Activists’ private jets in the air….;-)

    In terms of urban populations I suspect it would be more dial-an-electric-driverless-cab, which takes you to the nearest train station.

    Ships would be nuclear powered and fast.

    Aircraft would be tremendously expensive.

    But the need to travel; would be greatly reduced: teleworking of all sorts plus advanced robotics would mean that even hands on work could be performed remotely.

    • Willem Post says:


      Here is the land area to replace fossil fuels with bio fuels


      For the past 200 years, we have multiplied from 1 billion in 1800 to 7.4 billion in 2016 and most of them have lived the good life, courtesy of our fossil fuel inheritance. Transitioning to biofuels is like having to get a real job and work for the annual yield, year after year; a yield subject to the vagaries of the weather. No more freebees from nature that were just lying around for millions of years waiting to be scooped up. Luckily, we have about 100 to 150 years of fossil fuels left over.

      Eliminating fossil fuels from electricity production is one thing (expensive, but not difficult), it is quite another to eliminate them from the thousands of industrial processes that use fossil feedstock, and from the millions of products that consist of fossil feedstock, such as plastics, drugs, etc.

      The fossil fuel substitution issue is usually ignored by most RE proponents, because they are focused on gathering the generous subsidies to build mostly wind and solar systems, using the slogans of “getting rid of these evil fossil fuels, saving the world, fighting global warming, reducing climate change”.

      Corn Kernels Replacing Fossil Fuels: During the past 15 years, US corn production has become very efficient with large energy yields per acre. Harvesting corn requires modern machinery, good soils, fertilizers, various chemicals and adequate rainfall.

      Annual corn energy yield in the US: 160 bu/acre/y x 56 lb/bu x 7000 Btu/lb x 0.85 = 53,312,000 net Btu/acre/y, equivalent to 9.6 barrel of oil/acre/y.

      World Fossil Fuel Production Replaced by Corn Kernels

      Crude Oil = 80 million/d x 365 d/y = 29,200 million barrel/y.
      Land area in corn = 29,200 million/9.6 = 3.040 b acre.

      Coal = 8 billion metric ton/y.
      Land area in corn = (8 b metric ton/d x 27,778,244 Btu/metric ton)/53,312,000 net Btu/acre/y = 4.168 b acre.

      Natural Gas = 3,600 billion cubic meter/y.
      Land area in corn = (3,600 bm3/y x 35,315 Btu/m3)/53,312,000 net Btu/acre/y = 2.385 b acre.

      Total land area about 3.040, crude oil + 4.168, coal + 2.385, natural gas = 9.593 billion acre.

      World Land Area For Food Production = 18,963,881 sq mi x 640 acre/sq mi = 12.140 b acre, of which 28%, 3.40 b acre, is in annual crop production. For comparison, the US has about 0.434 b acres of cropland.

      If the world population were reduced to 1.0 billion, plus the energy consumption per capita were reduced by 50%, then 2.688 billion acres would be required for food and replacing fossil fuels with corn kernels. That is much less than the 21.733 billion acres that would be required, and even more for a greater population by 2050, if business-as-usual conditions prevailed.

      The 0.761 value likely is significantly understated, because crude oil, coal and gas require much less energy to produce thousands of chemicals for many purposes, than would be required to produce ethanol, etc., from corn kernels.

      Population reduction, increased efficiency, and FFs replaced by corn kernels would:

      – Allow the fauna and flora to reestablish themselves on 12.140 – 2.688 = 9.452 b acres.
      – More likely enable the fauna and flora to survive and thrive elsewhere in the world.

    • Any thoughts on the desire to travel, rather than the ‘need’ to travel?

      • Wookey says:

        People like travel, and it’s even good for them, and the world, to some degree. But we don’t have a choice about reducing GHGs if we want to preserve the civilisation in which that travel occurs in the long term, so travel has to either be low-carbon, or not happen.

        In the short term (10-15 years), that means significantly restricted travel, and modal shift (some modes have 1/10th the footprint of others). In the longer term different infrastructure.

        A great deal of modern travel is not very necessary, so I reckon most people could get used to doing rather less of it. They might even enjoy it.

        • William Forrest says:

          Once upon a time people traveled by rail, sailing ship, even horseback. The “convenience” of fast air travel is just that, a convenience.

        • Between 1990 and 2005 I drove 40 miles a day between my home in Tucson, Arizona and my workplace, burning about 2 US gallons a day in the process.

          Then in 2005 I “retired” to Mexico but kept working, doing exactly the same kind of work through a computer link. I burned no gasoline and the work I did was just as good (or bad) as what I would have done had I been in the office.

          And for the last few years I’ve been collaborating with Euan Mearns on Energy Matters without me ever going to Aberdeen or he coming to Mexico.

          There are of course many people who have to go to their places of work to do their jobs, but there are also a large number who can do it by telecommuting from home. I often wonder why more emphasis isn’t placed on this as a fuel-saving measure.

      • Leo Smith says:

        Well it used to be fun to travel, because where you went was different to where you started.

        Now there’s a Holiday Inn, a McDonalds and Coca Cola wherever you go.

        I suspect that in fact in a somewhat dystopian future masses of people will spend their lives plugged into a virtual reality that is infinitely more satisfying than the dull grey grimy council flat they find themselves actually inhabiting…

  11. Lars says:

    “The Guardian has no doubt that it will, “the EU initiative …. would open the door to a futuristic world in which cars supply energy to Europe’s power network at all times of the day and night, balancing shortfalls from intermittent renewable energies when the sun is not shining and the wind not blowing.”

    This claim is simply not true, at least not with present battery technology. Tesla`s CEO of development Jeffrey B. Straubel recently said that batteries in EVs are unsuitable for this task of balancing the grid because they are not designed for this. He is virtually shooting the whole idea down.

    Here is what he said word by word in a Norwegian article, my translation and comments in brackets:

    “The extra use of (EV) batteries (to balance the grid) implies a high price, and lots of people who think this feasible today don`t look at the wear and tear of the batteries, and the extra costs of having a two-way electrical system (to and from the grid). One needs a connection similar to a large PV installation and more costly overall standards he claims.”

    “The problem is that batteries designed for EVs are not made for a large number of cycles required to balance the grid.”

    So the question is, who the hell wants to send power back to the grid to earn a few cents while simultaneously ruining one precious charging cycle in a battery at a price of tens of thousands of dollars? None with a sane mind would do that.
    Perhaps someone with expert knowledge on batteries could comment on this.

    • gweberbv says:


      if there is a significant impact due to EVs on grid balancing then it will play out once the cars reach the end of life while their batteries are still working. These batteries may end up in utility-size installations for frequency control where they enjoy their second life with depth of discharge limited to – maybe – 20%.

      • Lars says:

        gweberbv, then we are talking about two different things, apples and oranges. I was referring to the fact that quite a lot of people (including the Guardian) seem to think that EV batteries can be used to send power back to the grid, which they can be technically of course at a cost of battery life. Would you do it under the circumstances Straubel describes?

        What EV batteries CAN do (cheaply) on the other hand is to create a demand response to the grid when there is surplus electricity and vice versa.

        • gweberbv says:


          I would not rule out the option that EVs sending electricity back to the grid completely. But I also do not expect this to happen. Simply because IF EVs become mainstream, THEN about 10 years later one will have a lot of used but still working batteries that can be used in utility-sized frequency control/demand shaping installations. So, there is no need to invent complicated regulations and payment schemes to utilize the batteries in EVs.

          • Lars says:

            gweberbv, it may or may not be. We don`t know exactly what the future holds. That`s why I read blogs like Energy Matters to see what people a lot more informed and educated about the subject than me think. But EV batteries` power to grid looks like a dead end to me unless there is some new revolutionary changes to battery chemistry etc.

            Personally my gut feeling is that electric cars are the future but not in the form of battery EVs but rather hydrogen cars. In 5-10 years time we will see which technology becomes major or if different technologies prevail in different markets and regions. I really hope that EM will cover this subject in the future and what prompts the japanese to go for hydrogen.

          • A C Osborn says:

            Lars, I am not sure why you think it is just Japan going for hydrogen, or for that matter Fuel cells.
            I worked for Ford Motor Co and retired well over 10 years ago. They has been working on Fuel Cell technology amd others for at least 10 years before that.
            How many Fuel cell cars do we see in anything other than prototype guise?

            I am not sure how they think they can overcome Hydrogen Embrittlement issues in Mass Production, Infrastructure and use.

        • Leo Smith says:

          What EV batteries CAN do (cheaply) on the other hand is to create a demand response to the grid when there is surplus electricity and vice versa.

          “Traffic is heavy on all roads today as the availability of wind power yesterday meant that cars are all charged up again and peole have to make their vital trips when they have the electricity to do so”

          I might actually believe what you said IF it were the case that the car battery contained significantly more energy than is needed for say a calm weeks motoring.

          But the reality is that EV batteries are at best uber marginal on capacity – cost and weight dictate that – and daily charging at least, is mandatory, and that will continue to be the case in the foreseeable future.

          EVs are not likely to be a match for regular hydrocarbon fuelled cars for a long long time, if ever. They are about a factor of ten out in terms of range right now, and there is not that much gain to be had in any battery technology of an electrochemical nature. Electro chemistry is not likely to change either.

          Any technique that turned out to be able to store large quantities of electricity at knock down prices and in complete safety and at low mass would of course change the world, and makes its developers a fortune.

          That of course is why everyone is crying out for subsidies and research grants. Such a technology doesn’t exist, nor has anyone even the faintest idea what it might consist of. But you can make a lot of money pretending that you do.

        • singletonengineer says:

          Batteries have limited life. Essentially, the equivalent of 500 full cycles tops off a lithium battery. Or a thousand @ 50%, or 2000 @ 25%. Plus or minus fudge factors for the effects of high rate of charge or discharge, high or low operating temps, etc.

          Those who envision a future powered by dead and dying ex-vehicular batteries are wasting their time and ours unless they first explain how they expect to resurrect the dead.

    • Wookey says:

      “The problem is that batteries designed for EVs are not made for a large number of cycles required to balance the grid.”

      I don’t really follow this. EV batteries are designed for daily (part) cycling – that’s how most owners currently use them. It’s not clear exactly how much extra cycling would occur from balancing. I’d expect it largely to take the form of delaying charging to a different part of the night (or day). That’s no extra cycling at all. On occasion there will be extra discharging, but it won’t happen every day, and for most users who still want some usable mileage at short notice discharge will not be deep.

      For people who don’t use their car much and leave it plugged in for grid support, I’d expect one cycle/day, same as if they’d been driving it.

      So will there really be a lot of extra cycling at levels that cause significant extra battery wear? It’s not obvious that this is the case.

      And yes, of course, none of this will happen unless it’s required, or there are adequate incentives in place to cover the battery wear.

      It’s worth remembering that cars spend 96% of their time parked, so they are widely available for grid support, at least in principle. Cars are moving (on average) for about 1hr/day, and parked at home/work/somewhere the other 23 hours. Assuming charge connectivity at home and work and some car parks that’s probably still 22hrs/day average (potential) availability.

      I actually think that what’s more likely is using retired automotive batteries for grid storage, because there will (in a few years) be an _awful_ lot of these and it’s a lot easier to just plug those in than worry about complicated protocols and pricing models for dual-usage of car batteries still in cars.

      • singletonengineer says:

        As referenced immediately above, the most that can be expected from stationary house batteries undergoing partial (say 25%) cycling is a couple of thousand cycles… say, 6 years. Other references give 3 years as the practical life of lithium batteries, by which time their use as energy stores is minimal.

        In a few years, there might as claimed be an -awful – lot of batteries available that were formerly in EV’s, but will they be worth a crumpet?

        My guess is that, similar to discarded automobile tyres, they will generally just pile up in millions of locations around the globe as waste, despite there being hypothetical recycling avenues available.

    • Lars,

      Kan du vise oss en link til artikkelen? Takk!

  12. Jesús de Lucas says:

    Thank you. Simple and clarifier. For sceptics, this nice analysis has been done on current premisas and knowlege. Tomorrow, probably some technological aspects will change, but no will change the adress, but yes, it will change the speed to change the current model… Typical dilema, pionners, followers and the conservatives.

  13. Joe Public says:

    3 years ago, Paul Homewood wrote a UK-specific post “Electric Cars Will Create £14bn Black Hole For UK Government”

    It highlighted how much petrol/diesel motorists contribute towards Treasury coffers.

  14. Alex says:

    Roger, I’ve been doing the same analysis for the UK. The data I took was from here:

    You gave the number of cars but the mileage figure is for All vehicles. In our assumption we assumed cars are 90% electric and HGVs 20%.

    The average UK car does about 16,000km, which is only about 300km/week, which is about the range expected for the Bolt. So the average car will need about 50KWh per week, and could charge twice per week.

    It’s reasonable to assume most cars will be plugged in when at home – most EV owners with off road parking do that – but on street and communal parking facilities need a massive upgrade.

    I’m very sceptical about V2G for a few reasons, but controlled time of day charging could be used to switch demand towards night time and towards weekends (vehicles drive less at weekends, hence they’re more likely to be plugged in). You don’t have to control all vehicles all of the time to make a change.

    This is much easier to control with a steady supply – such as nuclear. People get into habits – “I charge at night”. It’s harder with a random supply. “I alter my routine to charge when the wind blows in the North Sea”!

    • meliorismnow says:

      Mass adoption of EVs and TOU plans can level out demand. No new infrastructure is needed to increase the demand at nadir (and off peak) to that of the peak if the generator is stable (nuclear, coal, NG). It will actually decrease costs/kwh because of the higher utility of said infrastructure.

      If you want to add renewables, you have multiple options which you probably want to combine:
      a) ramp legacy generators based on intermittent supply
      b) curtail unexpected peaks of renewable sources. This generally should be done at least slightly so you don’t need distribution infrastructure to deal with surges that will almost never be present.
      c) Amplified TOU plans (seasonal, hourly, and instantaneous)
      d) export at peaks and import at nadir
      e) utilize hydro as a “free battery”

      With mass adoption of EVs:
      1) you can have fleet agreements with uber/lyft etc where autonomous EVs park and plug into RE assets absorbing most if not all curtailed energy near to metro areas (or highways/rail for shipping). They’d also take low value energy (unexpectedly high supplies during unexpectedly low demands for the local grid). The charging could all be HVDC (as currently used for L3 EV charging) with little to no conversion losses. Autonomous EV taxis also greatly reduce the need for vehicle ownership and greatly increase carpooling amongst the poor and middle class (some at the expense of mass transit). Yes, a larger portion of their fleet would be idle but after reducing operational costs by an order of magnitude (driver, efficiency, and transmission, motor, and brake longevity) electricity will be a major cost (along with financing).

      2) If an EV has a 60kwh battery but only uses 8kwh on 95% of days with a known schedule, demand and V2G supply can not only shift intra day but intraweek. Smart meters can be integrated directly with the charger with very little added cost and wired the same as solar systems are currently (except with some additional internet connection; either RF cell tower or personal WIFI). They could take advantage of instantaneous power pricing (whether visible to the meter or controlled by the utility). This could even take into account transformer, line, and substation health and capacity, increasing the utility of the current infrastructure.

      3) By the time you reach 100% adoption, you probably have at least half that capacity of used cells that can be reused in stationary applications. Manufacturers like Tesla could actually reuse both the pack and the BMS, possibly with some cell swapping (which could be automated).

      • singletonengineer says:

        Has this comment overlooked the small fact that, when storms damage powerlines and thus load is lost, what is required to keep the system stable is a combination of more synchronous generation (to maintain frequency as loads ramp fast) plus more available load (ie, 10 to 30% full and thus able to accept the surplus for a short time while generators adjust their output).

        So, the batteries have to be full and empty at the same time – some trick!

        In common parlance, “spinning reserve” works both ways, with the equivalent of one or two times the size of the largest thermal generator in the system online, able to ramp up for more load. Of course, all thermal generators are able to ramp down, at least for a time – even the so-called baseload nuclear stations can do so.

        The battery folk need to talk of a thousand MW availability both ways,for as long as it takes to start standby generators from cold – at least 30 minutes and perhaps more than 30 hours.

        A design for a 60GWH of batteries (50% charged when under stable load) would be a good start, but who is going to find the money?

  15. clivebest says:

    There is not enough Lithium in the world to make this scenario feasible. Manufacturing 20 million electric cars a year in Europe would need about 250,000 tons of Lithium per year. This would exhaust world supplies within 50 years. However if you include USA, China and Asia then Lithium supplies would only last 10 years.

    We actually need Kryptonite!

    • See the posts from Beamspot, Robertoko6 and my inferior posts from Blowout 146. Reserve is around 13 million tons and reserve is just over 34 million tons.

      Current world wide production is approx 36,000 tons per year.

  16. roberthargraves says:

    I didn’t realize the Bundesrat was thinking of autos as energy storage devices. I thought the idea was simply to reduce petroleum consumption.

    Willem Post’s Tesla experience (.38 kWh/mile = 24 kWh/100 km) is close to your assumption.

    I plugged your numbers into this table to ballpark the number of 1 GW ThorCon power plants that would be required to service the global fleet of electric cars, simply assuming 100% uptime and round-the-clock even recharging load.

    Cars (millions) GWh/year GW
    Germany 44 203189 23
    Netherlands 8 23414 3
    Norway 2.5 9746 1
    EU 250 1100000 126
    UK 25 122386 14
    US 261 1229925 140
    China 154 5810000 663
    World 1000 4400000 502

    We’ll plan on $2×502 billion capital expenditures for ThorCon power plants. (:-) Figure 3-cent/kWh generated electricity will sell for 6 cents at the charging station, so plan on $0.06×24=$1.44/100 km.

  17. halken says:

    Using cars as grid batteries is highly questionable, as it will depend on the price premium the grid will pay for this, and even if it works, then it cannot provide energy storage beyond a day or two. Here in Northern Europe we can have a high pressure system dominating for weeks on end in the winter, and that spell disaster for such a system in a few days. Other energy sources is needed, such as nuclear and preferably thorium MSRs on a large scale.

    In this video, with the guys who made it in the 1960 and some scientists from today, they asses that building a demonstrator plant in EU will cost 10 b€. Compared the net gain such a technology could provide, one might wonder why we’re not already investing in R&D for this technology, that can power the EU without oil/coal, without costly wind turbines and solar panels in nature and grid reinforcement and backup power plants, without nuclear proliferation, without nuclear accidents and without nuclear waste.

  18. Olav says:

    Peak Load from EV is challenging. We will need fast chargers that refuses to charge when grid is overloaded. Telling you… Expected time for charging to start is 2h 12 minutes.. You may consider to move over to the nearby “slow charger”..

    Besides that is my personal experience I Norway with EV extremely positive. Long range for me is 160 km and then I Speed Charge for 1£ after 120 km in 5 minutes. Max range for my car is 150 km
    Signals are given that purchase subsidies are phased out in near future but toll cost will always be at 50% of fossil fuel equaling car. This is a budget “haggling” issue so a promised green “energy wende” may change things in favor of EVs.
    Nevertheless the big thing is the low cost of electric traction vs fossils. Driving at 25% or less cost will persist so EV is here to stay.

  19. Euan Mearns says:

    How much Lithium?

    In the last Blowout OpenSourceElectricity said this:

    In one average km³ of earth is 120.000t of lithium. There is no problem to ramp up lithium production, it is not rare, it’s just not concentrated to high percentages by natural processes anywhere.

    Lets check that sum.

    Average density of crust ~ 2.8 so 1 m^3 weighs 2.8 tonnes
    Average Li ppm in crust ~ 20 (ppm = g / tonne)

    1 km^3 = 1*10^9 m^3 = 2.8*10^9 tonnes = 56*10^9 g of Li in 1 km^3 of crust
    = 56,000 tonnes of Li in each cubic km

    I have another source that says 56 ppm Li in the crust, so open source was on the money. The trouble is this is disseminated throughout the rock. I’ll let someone else workout how much energy is used to mine, grind and chemically process 1 km^3 – its the sort of number Roger probably has in his head 🙂

    All minerals / metals on Earth are mined where geological processes have enriched the target metal so that it can be economically mined.

    According to Wikipedia, global production is running at 600,000 tonnes / year. And global resources are about 40 million tonnes for an ROP (reserves over production) = 67 years. Production of course will not stay constant or rise linearly. It will follow a Hubbert type curve with rapid rise in the coming decades followed by a fall as the easy to access resources get used.

    But I wouldn’t bet against new deposits being found.

    (if there’s a problem with my sum let me know. I’ve done it three times and got a different answer each time)

    • Alex says:

      As with Uranium, Lithium can be extracted from sea water. The question is cost, and that provides an upper bound to Lithium prices.

      This graph shows how near different minerals are to being viable for sea water extraction:
      So for example, lithium is almost worthwhile now. Uranium has to triple in price, and silver is impossible, so solar panels will have to find alternatives.

      I also think that for buses and large vehicles, sodium batteries may work out better than Lithium.

      • Beaspot says:

        See my coment above regarding resources and reserves. USGS is the most credible source of that kind of data.

        Regarding the extraction from sea, well it will require to process like 5 Nile rivers, at 170ppm’s. I guess any guess regarding this concept is almost BS.

        Gold is more common (>200ppm’s) and yelds much more benefit at much higher price per ounce than lithium, but it is still to be seen that it is extracted from sea.

        Besides that, Lithium is used to reduce the cost of PV glass about 15% (by reducing the amount of natgas to produce it), so if Lithium cost rises, many other things will also rise.

        • Alex says:

          Here’s the link.

          According to the chart, Gold is several orders of magnitude away from being economic.

          The Nile doesn’t have much flow.

      • OpenSourceElectricity says:

        Solar panels today require still less silver than Photographs did some years ago.
        Lithium can be produced in high volumes if the price is high enough to process Pergamatit. Which is true today, so many new mines open. Producing it from brine is cheaper today, but if demand is high econmy of scale might bring down prices producing it from rock too.

    • Sorry Euan, but I don’t have that number in my head.

      The problem with extracting lithium from rock is that it occurs in economic concentrations only in pegmatites that contain spodumene (LiAl(Si2O6)), and there aren’t many of these deposits. That’s why the world’s lithium now comes dominantly from lithium-enriched brines in enclosed drainage basins. The basic question is therefore how much lithium is contained in these brines, not how much is contained in the upper crust. I’ll take a look at this as time permits.

    • Euan

      The wikipedia production rate is nonsense. That would mean not even 2 decades of reserves. The USGS that wiki references estimate 32000 tons pa.

    • OpenSourceElectricity says:

      The mayor point is for me: if there is a higher demand and so higher prices of Lithium per TOn, the amount of resources where it can come from multiplies. Also the likelyhood of recycling lithium.
      Higher prices and higher market volumes also lead to more spendings in research how to extract Lithium from other sources. So far the marktet for Lithium was very uninteresting.

  20. Roger

    Regarding subsidies in Norway, it current stands (all party agreement, dates may vary a bit)

    Owners will have to pay half of the road tax from 2018 and all of it from 2020. VAT exemptions will also be replaced by a subsidy, which will probably be revoked over time. And local authorities can now decide whether to offer free parking, exemption from tolls, and the use of bus lanes.

  21. Wookey says:

    Where do you get 15% charging loss from? Having measured my own EV (a Scooter/Moped). The charging efficiency was 94%. i.e for each kWh taken out of the battery for the motor, I took 1.06 kWh from the mains. So that covers charger inefficiency and battery cycle losses. This was for lithium iron phosphate batteries (and daily cycling), but the characteristic for Lithium-ion or lithium-polymer is very similar.

    More expensive chargers would improve efficiency. Fast charging would decrease it, but neither would get down as far as 15% losses.

    • meliorismnow says:

      Exactly, 94% is pretty standard for L2 charging. L1 can be higher or lower depending on the adaptor&BMS and how you measure (Tesla’s are constantly using power for thermal control and wireless communication so they appear less efficient on low charging rates because a greater % goes to those systems). L3 are certainly less efficient on average but there are plenty of R&D left to go on those and in theory they can be implemented to avoid conversions they currently go through (say paired with a solarcity PV system & tesla powerwall).

    • meliorismnow says:

      Forgot my link showing typical 94% and achieved 97% charging efficiency:

  22. Hugh Sharman says:

    Roger, you really are a saint! Thanks again!

  23. Phil Reed says:

    Once again someone misses the elephant in the room.
    For every gallon of petrol or diesel produced the refinery uses 7 to 8 kWh of electricity.
    The change to BEV’s will allow the large amount of Gwh used for refining to be used to charge cars.
    You need to work out gallons of fuel that will not be needed then factor in the electricity saved.

    • roberthargraves says:

      So much electricity to run refinery?! 7-8 kWh? Generating that 8 kWh(e) electricity from coal probably consumed 20 kWh (thermal). A gallon of gasoline is about 45 kWh (thermal). EROI of only 6:1, excluding the petroleum?

      • gweberbv says:

        I found a similar number here:
        However, to me this does not look like the refinery is really using this amount of energy exclusively in form of electricity.

      • Alex says:

        Robert, FYI, nuclear heat at 600C can take the chemical efficiency of steam methane reforming (thermal energy of hydrogen / thermal energy of methane) from 70% to 120%.

        That means, you could “turbo charge” a MSR, effectively trebling it’s output. Or you could make hydrogen production viable.

        Might be of interest if you know of any Molten Salt Reactor developers.

        • roberthargraves says:

          ThorCon does have the heat to do this. There are four heat transfer loops: fuel salt, clean salt, solar salt, steam. The steam temperature is 538C, normally used to drive a steam turbine-generator. The solar salt temp is 598C, and the nonradioactive (K/Na NO3) salt could instead be pumped through chemical reactors in a refinery. Such a fission power plant would be cheaper than a full electric generation plant; figure $500,000/MWe saved by not buying a steam turbine-generator. I bet there are many other processes that clever chemical engineers could devise using cheap heat.

    • It doesn't add up... says:

      I think you must be confusing things: a complex refinery will consume perhaps 7-8% of its crude oil input in process energy, most of which is in the form of heat. Most such refineries have their own onsite generation (they also have uses for steam to keep pipes and tanks warm, and for cracking to hydrogen for chemical processes): a quick estimate is that a typical refinery would use comfortably less than 0.5kWh/gallon as electricity.

  24. Hans Erren says:

    Just like we don’t mine ice anymore to run our refrigerators, the future will be diesel fuel synthesized in thorium or fusion reactors. The big issue of electric cars is electric charging logistics.

  25. A C Osborn says:

    Roger you stated for item 3.
    “I used the median value of 22kWh (equivalent to 153 miles per US gallon according to the US EPA) obtained from 114 test values published by the US DOE.”

    Chefio (E M Smith) has raised serious doubt over the voracity of that 153mpg figure.
    See here

    • AC. Smith may well be right, but it doesn’t make any difference to my calculations.

      But if he is right we would probably do better to concentrate on hybrids, which don’t need to be plugged in to the grid and which will ultimately be capable of delivering an average of maybe 50mpg US. Or will they?

      • ristvan says:

        RA, for many reasons published by myself elsewhere you are likely correct about full hybrids. I have driven a full hybrid Ford Escape 4WD Small SUV since mid 2007. We still get 32 city/28 hwy at 70 mph after ~70k miles, despite AWD and a class 1 tow hitch. We pick up 15% mpg because of the Atkinson cycle engine rather than Otto cycle. Lower Atkinson torque is compensated fully by the 82 hp electric machine. We pick up another ~15% because a 120hp I4 engine rather than the directly comparable 200 hpV6. We pick up 5-7% city because of idle off at stop. And we pick up 12-15% city thanks to 85% regeneratve braking. But the best part is the Atkinson I4 runs on regular gas. The comparable net hp (~200) Escape Otto V6 runs on premium. The price difference is right now about $1/gallon.
        Hybrid premium paid for itself instantly given the then hybrid tax rebate. Paid for itself in two years ignoring rebate given fuel savings and premium difference. After 9 years, the whole car cost is more than ‘free’. Best part is, hybrid Escape taxis are logging over 300000 miles in NYC with no evidence of battery deterioration. We expect to see 15 years on our vehicle as driven before any signs of battery deterioration. We will be fighting rust by then in our seashore climate in South Florida.
        Only ‘sacrifices’ were ~4 inches of trunk space depth (no matter in an SUV), plus 350 extra battery pounds fully compensated by superb electric machine torque on acceleration. We can still squeal tires.

  26. Syltty says:

    Tesla CTO JB Straubel On Why EVs Selling Electricity To The Grid Is Not As Swell As It Sounds

    ““If we want to actually send energy back from the car to the electricity grid, this gets much more complex, and, you know, that’s something that I don’t see being a very economic or viable solution — perhaps ever, but certainly not in the near term. You know, the additional wear and tear and degradation on your vehicle battery has a fairly high cost, and many of the people and small businesses looking at this today, you know, don’t take into account fully that degradation cost, and also the additional interconnection cost, because if you interconnect your vehicle, you do have regulations that play a part — it has to interconnect in the same way that a solar system would on someone’s home or on a business, which have different standards so that they can protect line operators and people on the grid.””

  27. ristvan says:

    Roger, I think your analysis understates the additional generating capacity needed by up to a factor of ~2. Take the Chevy Volt PHEV as a first illustrative example. Volt now has a 17.1KWh battery, with an effective (temperature dependent) battery only range of 38 miles. The range was chosen because about 2/3 of US urban/suburban car trips are under 40 miles. That battery takes 4.5 hours to charge at 240V, and 13 hours at 120V. At 240 V the car could be charged twice in a day, say while at work and then at home at night. At 120V, mostly only once at night. In either case, it is clear that the bulk of charging will be at night. Solar is useless then.
    Now it is also true that in both UK and Germany peak load is winter evenings. So charging at night in winter is on top of the preexisting peak load. You cannot average a day’s or a year’s requirement just looking at the additional generation needed on average. It is the incremental EV charging load on top of the peak that determines the requisite additional capacity.
    The reason this really bites is the charging time. Second example comfirmingbthis is the BEV Tesla S. Standard battery is 60KWh with advertised range of 232 miles (EPA says actually only 208 because of temperature dependency. Using a 240V system it recharges 31 miles of range per hour of charge. So full charge takes 7.5 hours. Half charge takes just under 4 hours. With 120 volt, the rate is only 5 miles/charging hour and the car would be plugged in all night (12 hours) just to get about 55 miles of range. Both are Volt charging time similar. Again, the charging load is mainly at night and the winter peak is the floor from which additional capacity has to be added, sufficient for ~12 hours.

    Reason it could be less than 2x your estimate (created by 12 rather than 24 hours of likely charging) is that the more 240v is used rather than 120, the more that peak can be cut 2/3 by spreading three separate 4 hour charges into the 12 hour period, then dropping the first period during the true peak, and just have 2 4 hour charging sessions after 10pm, assigning all odd serial numbers one 240V time slot, and all even serial numbers the other.

    • Alex says:

      “So full charge takes 7.5 hours.”

      Do you know what the “7” in “Economy-7” stands for?

      • Grant says:

        If the “Economy” of cheaper electricity base load overnight becomes, as seems likely in an electric centric car world, peak charge time … there will be no “Economy x” as we know it.

        If overnight electric charging does not happen then a large part of whatever ‘real’ benefit might be linked to electric vehicles becomes questionable based on existing historic base load concepts.

        On the other hand make the connection to solar generation and charging during the day might start to look sensible in some parts of the world.

    • Willem Post says:


      You are quite right regarding when charging takes place.

      One has to assume charging patterns and have generation to match the EV demand, on top of normal demand patterns.

      Normal demand patterns could be modified with time of day rate schedules, but the EV demand would be a very significant addition.

      That kind of simulation would be rather complex going from US coast to coast.

      I don’t know if that was ever done for 100% traditional generation, and a future steadily decreasing traditional and steadily increasing sun/wind-dependent generation.

    • gweberbv says:


      at least in Germany the lowest demand is always between 11 p. m and 5 a. m. During these hours you can add up to 15 GW of constant demand without problems (this is roughly the difference to the demand during daytime).
      See here:

      What might be more of a problem is the fast charging option. Most houses in Germany have a grid connection of 40 to 60 kW. For appartments it is still about 30 kW each. The grid will have big a problem, once a lot of people start charging their car around the same time with something like 30 kW each. But this is just a question of regulations/pricing schemes.

      • Dave Ward says:

        “Most houses in Germany have a grid connection of 40 to 60 kW. For apartments it is still about 30 kW each”

        Are you sure about that? It converts to about 130-260 AMPS, @ 230volts. That is WAY more than any domestic property in the UK. Most houses have typically had a supply fused at 60 amps, and newer ones are now 100 amps. Even my friends commercial chicken farm only has a 3 phase, 100 amp per phase supply…

        • gweberbv says:


          I just looked it up: The standard for a single family home is 3 phase with 40 amps each if no electric heating (water or space heating) is planned.

        • OpenSourceElectricity says:

          Yes, in germany standard fuse for 1 family homes is 3x63A, only some old connections here have 3x35A. In my house wiring also allows 3x100A. which would be 69kW in my house.

          • Dave Ward says:

            Thanks both of you. I’m surprised! Few domestic properties here have 3 phase supplies as standard.

      • Alex says:

        “During these hours you can add up to 15 GW of constant demand without problems”

        As long as the lignite plants are going. If you have a steady source, it’s easy.

        “The grid will have big a problem, once a lot of people start charging their car around the same time with something like 30 kW each. ”

        There could be local problems at sub station level. National capacity is designed to supply industry and homes. If at night industry is quiet, but homes have doubled their demand, local substations will be over loaded.

        But 60 hours@30KW is 180KWh, which is enough for 1,000km of driving. Not every house is going to download 180KWh on the same night. (Unless the weather forecast says no wind – reservoirs low – panic buy electricity).

  28. William Dunn says:

    At the risk of asking the obvious – Withall these cars acting as storage during times of low wind/solar production (mostly night/winter/cloudy)- could mean low/dead batteries to power the car next morning ? – not going to be too many pleased with that .

  29. Philip says:

    EV vehicles when they arrive (presently a tiny % of total) will not replace present day vehicles. It is expected that with Autonomous driving systems car ownership will plummet and people will use what ever replaces UBER, in cities at least. Thus trying to calculate EV power requirements based on how we use cars today is a waste of effort. Autonomous driving vehicles are the real disrupter to transport not EV.

  30. beamspot says:


    There are many quirks regarding EV’s and electricity that have to be considered.

    First of all, there is the issue with battery life. There are two main ways for battery aging. The first is the most known one, battery cycling.

    For Li ion batteries, the product of cycles times depth of discharge is pretty constant from 5 to 80% DoD (Depth of Discharge). Above that, life shortens, but not significantly, unless you deep discharge them below2.8V per cell. You wouldn’t overcharge them unless you are a Samsung Galaxy Note 7.

    One year and half ago, I’ve said that Tesla cells will last between 800 and 1000 100%DoD equivalent as a best case. Now that Tesla is publishing data, it seems that this value is in the 700 to 750 cycles.

    That is interesting because that allows to calculate ESOI (or Energy Stored on Investment). For Tesla Model S80, it seems that those batteries last about 300.000Km, and the battery of 65KWh costs about 30.000$, standing for 10$/100Km, about the double of the cost of Diesel of my 407 2.0HDi. Same approach can be used for Power wall, with about 7MWh of over all stored energy (for the 7KWh unit, the only remaining in the marked, at 1000 cycles 100%DoD, optimistic estimation), that for 3000$ stands at about 0.43€/KWh, plus other costs that will dirve up costs per KWh extracted from the unit well above 0.5 or even 0.7E/KWh.

    The second main drive for aging is temperature related, or Arrhenius speed of chemical reaction. That one is funny. Warmer countries see the batteries last less as temperature rise than in colder countries. The result is an imaginary parallel by the Pirenees that reduces battery life south of that line to 5 years or less (as you go south) whatever you do with the battery.

    That is the reason why VolksWagen told me off the record that Spaniards will never enjoy an Electric Seat (pun intended). You can check for Electric Seat Leon (the counterpartner of VW Golf) that will never be sold, because thermal aging of batteries doesn’t not encourage to sell BEV’s south of that imaginary line. Curiously enough, those are exactly the PIGS countries. Just those that, given the fact that they are poorer, they are less likely to buy an expensive electric vehicle, so even easier for car manufacturers.

    That also puts an extra and overlooked pressure on thermal management for pack batteries, that implies oversized heat pumps, or event who of those units, where thermal management of batteries take precedence over anything else, even range.

    Thus, in cold climates some battery warming is performed, but their use also warms them, and Li chemistry works fine at low temperatures. So the real problem appears in hot climates, like north of Africa, or Seville in august, 45ºC mean, under the full sun. That will pose a big deal to battery life, so air conditioner starts at about 30 to 35ºC to keep batteries below 30ºC.

    Those systems can draw up to 7KW for large packs, like Tesla Model S.

    Thermal management, specially for inefficient, cylindrical worst geometry possible for cooling, high energy (thus low power) batteries used by Tesla impose an expensive and heavy liquid cooling schema, rising costs even further.

    While mass production of battery pack may drive the cost of assembling them down, batteries are already manufactured in millions of them per year, so no substantial improvements can be expected in that field, but, even worse, the increase in demand is driving cost of raw materials (that represent about 70 to 80% of the cost of the cell) to skyrocket. Lithium carbonate battery grade raised from about 6000$/T to more than 19000$/T. Cobalt is also raising.

    This cost hardly will be offseted by huge battery assembly plants.

    Car manufacturers already know that. All of them. That is the reason why, all of the ones that I’m aware, are developing plug-in serial hybrid schemes. For sure BMW, that is my main customer, but almost all of them (VW, Fiat, Renault, etc).

    The reason is simple: smaller power battery means higher probabilities to wear it out by cycling it within 5 years, lower cost, and no problems with large trips since the FF powered generator will take care of that, as a ‘safety backup’, while for the more common short range will be kept electric.

    Be aware that electric drive is particularly efficient at city drive, reason why all of them use the EPA City drive cycle instead of road ones, just the contrary than FF cars that are particularly good at road and highway, with really bad city behavior and consumption.

    So, IMHO, and, as it seems also the german multinational company where I work, the trend will be towards the use of small electric autonomous and cheap cars, like google car or similar, that will drive private cars to oblivion, and reduce car park by 90% (estimation of the automotive electronic car manufacturer where I work).

    There are many more point and implications to explore in this area, but now I have to leave. I hope I will have the time to add more points and considerations later.


    • gweberbv says:


      is it impracticable to put insulation around the batteries (in addition to active cooling/heating)? If I look at our fridge, it should be possible to keep a volume of a few 100 liter about 20 K cooler than the environment for less than 0.5 kWh per day. If the battery is fine with 25 degree Celsius, the need for temperature management does not look like a major issue to me.

      The only problem occurs when the car is disconnected from a plug or more than a few weeks.

      • Jonathan Madden says:


        Battery temperature regulation by whatever means has to compensate both for ambient temp and also for heat generated on battery discharge, from drawn current and internal battery resistance.

        Reversible heat pump, straight electrical resistive heating, and/or blown ambient air have to accommodate the worst case the car is specced for. A car from the southern European market will probably be designed for parking and use down to -20C outside as well as up to +40C. And vice versa for one sold in, e.g., Norway.

        To keep weight, complexity and electric power draw within reasonable limits for battery protection it seems wise to provide battery thermal insulation in order to keep cooling power down, for the norm of high ambient temp in a hot climate and to allow the heat pump to operate efficiently. But the cooling system has to be able to deal with the car running at high power for an extended period, in hot weather when simple fan cooling won’t be very effective, as well as cooling when parked.

        So battery insulation is probably standard in an EV, with a suitably powerful heat pump for high temperature/high power management.

        • gweberbv says:


          thank you for your comment. But now I do not understand why battery life should be significantly reduced in ‘warm countries’ as Beamspot calimed.
          Maybe one will probably have to use an extra 100 kWh per annum to cool the battery compared to a country like Denmark. But that’s it. Or isn’t it?

          • Jonathan Madden says:

            The electrodes are subject to degradation with increasing temperature of the charge/discharge cycles, the cathodes being more affected than the anodes. However all aspects of battery function deteriorate faster at elevated temp. Have a look at:


            for experiments in the 25-55C range.

          • Beamspot says:

            Yes, you are right. Thermal insulation is standard procedure in EV’s, and yes, they consume very litle if they are plugged.

            But that has two issues. One, is when it is no plugged, that it can drain all energy withing few days/weeks (depending on the country, environment, etc), but this is, in fact, the point with some applications (public city buses), but at some range cost (not an issue with public city buses anyway, but for more generic particular cars).

            The second is not as straight forward, but this is a big plus in two areas complexity/cost of battery pack assembly, and increased weight (that also means increased consumption).

            After all, the issues with batteries boil down to two issues: lower results than usually publicized (and often, by far), and higher costs, being the last one the key behind the EV2.0 failure (and I guess, also the point behind not only EV3.0 but peak oil/renewable energy failures).

  31. confused mike says:

    The analysis is clear and helpful.
    I also seem to recall that the strategy to reduce emissions from residential heating was also to switch, for instance here in the UK, from Gas fired to (clean) electric heating – presumably over the same time horizon as these EV initiatives.
    Is there a similar figure for this effect both nationally and across the EU of this impact on winter power generation sizing?
    I seem to recall that some countries like France are heavily into electric heating – supported by their nuclear power portfolio – but other countries I believe are similar to the UK (the Netherlands for instance) in being either gas fired heating individually or district heating schemes which eventually will be converted(?) if the emission goals remained unchanged.

    • gweberbv says:


      for heating the numbers are harder to crunch because it is impracticable to heat most of todays buildings with heat pumps. Thus, for heat pumps to become mainstream not only for new buildings but also for the older ones, you need significantly improve the insulation of the latters ones plus you would also like to install floor heating (or panel heating). But this would also reduce the need of heating.

      Bottom-line: The question of how much electricity would be necessary for decarbonisation of space heating boils down to an estimate of the heating requirements of a completely refurbished building stock. Hard to predict, I would say.

      • ristvan says:

        Heat pumps are fine for moderate climates like southern or Germany or Southern US. They do not work at all where winter climates are more severe, like upper US midwest. The only heat put out there is the pump motor heat loss. In such places, you would have to go to electric resistive heating. Assuming the best 61% CCGT and a standard 10% loss through T&D (transformer heating, line losses) then 100% natural gas heat value translates to 51% electrical heat. Makes no sense. My newest propane furnace at the farm is 95% edficient, and my new gas furnace in Chicago is 80%efficient. The only reason I did not buy a 95% for Chicago was we reused the old flue as is. 95% would have required an insulated flue liner, an extra exhaust blower beyond whatnis built into the furnace, and a moisture drain. At US natural gas prices thanks to fracking, more expensive than taking the 15% fuel hit.

      • Alex says:

        It would be quite straight forward to fit air to air heat pumps on most of the UK housing stock – certainly anything built from the 1980s. Outside temperatures are usually warm enough that a ground source is not needed.

        The main problem with Air to Air heat pumps is with draughty houses. If the house is too draughty, then there are more serious issues that need to be overcome.

  32. Pingback: How much more electricity do we need to go to 1...

  33. matthew_ says:

    The cost of the new capacity is a bit less than four times the current annual fuel cost for driving the same number of km. (Assumed consumption of 0.7 liters / 10 km
    An investment that has a payback time of less than four years doesn’t sound too bad.

    • robertok06 says:

      “An investment that has a payback time of less than four years doesn’t sound too bad.”

      Ahem… you forgot the payback time of the EV!… this is only the payback time of the power generators.

      • matthew_ says:

        The vehicle cost is not forgotten. The comparison is electric fuel vs fossil fuel for vehicle propulsion. The cost of an el-vehicle is similar to the cost of a fossil-vehicle.
        The main point was to show that the cost of the new capacity is quite low compared to what we pay today for our fossil fuel.

  34. PhilH says:

    The UK’s electricity consumption has been falling by more than 1% a year since 2005. If that were to continue, by the 2030s that would free up most of the current generation and generation capacity to power the EVs, so there wouldn’t be much more capacity needed.

    Obviously that’s a big if. Anyone care to estimate how much further scope there is for energy efficiency in electricity use? I’d guess that the replacement of incandescent bulbs by LEDs and CRTs by flat screens and more-efficient refrigeration & white goods and general building insulation has maybe run half its course, so there’s at least another 15% reduction in the pipeline of current technology alone.

    • singletonengineer says:

      That probably applies more to domestic and commercial loads than to industrial ones, for which achieving reductions might imply reduced production and hence unemployment.

      What percentage reduction in power consumption is achieved by installing LED globes in an electric train?

      I suggest that more than 2 sentences of analysis are needed before a 15% reduction pipeline can be assumed. Besides which, aren’t the low hanging fruit always the first to be eaten?

    • Leo Smith says:

      the new build next door is featuring air source heat pump. No gas and oil doesnt meet regulations apparently

      forget less electricity. make that more

  35. Jack says:

    Another problem with using cars as storage which I don’t think has been mentioned is timing. UK peak electricity demand is in the morning just before the comuter fleet sets off and in the evening just after it gets back.

    That being said, batteries do provide a technically feasible soulution to short term intermittency, albeit at a cost. But there is NO techology available which will deal with three windless winter days nor the summer-winter variuation of solar outside the tropics.

    Somebody suggested hydrogen powered vehices as an alternative to electric. I find this unconvincing. Round trip efficiency for hydrogen is less than 50%. Its physical properties make it the nearly the most inconvenient energy carrier one could devise.

  36. Pingback: Elektroautos: Deutschland sucht drei weitere Energiewenden

  37. Pingback: Effetto Risorse and collaterals 4 | Pearltrees

  38. robertok06 says:

    “(If the UK’s 25.8 million cars were all 30kWh EVs they would store 774GWh when fully charged, almost a hundred times as much as the Dinorwig pumped hydro plant.”

    Possibly true… so even assuming that ALL of the energy stored in them is available to do other things… this amount of energy falls SHORT of one day of electricity consumption in winter.

    I have tried to imagine myself in this fantasy world… and I swear that I can’t see how I would be willing to drive a car that needs to be parked at all times in order to fix the problems of OTHERS, and the same car need to be parked during the day to be charged up by PV, then during the night to be charged up by wind when PV is not available (most of the time, even in the future, unless the geniuses at The Guardian find a way to fix the sun’s orbit at zenith above London on 21/6….

    There’s only one way to fix the “environmental” problem of the planet in an affordable and sustainable way… and it starts with “n”.

    According to the data above, European reactors alone (810 TWh in 2015) could power NOW in a clean way 75% of the car fleet of the continent (1100 TWh)… why look elsewhere and try to re-invent the wheel????


  39. Neil says:

    Thanks for the analysis, I did a similar basic analysis and the figures aren’t too different, although I arrived a greater requirement for installed capacity.

    Assuming all 23.7M cars on British roads (2013 data) convert to 100% electric vehicles like the Renault Zoe, which is a small, modest car. This has a range of 60 miles and a battery capacity of 22 kW hours. Assuming all cars are the same spec and each is used for the full 60 miles/day:

    22kWh x 23.7M vehicles equals energy use of 521 GW hours/day.

    If charging all these cars takes place evenly over a 24 hour period, you need to allow peak power input of 21 GW.

    Assuming power stations have a 40% conversion efficiency (fuel to electricity) then you need extra capacity of 54 GW to charge all these cars. This compares to a current peak UK demand of ~55 GW

    If the grid isn’t smart and can’t manage to move car charging times to average out the necessary power over 24 hours, and we all charge over the same 4 hours every evening, then you need closer to 325GW!

    • Wookey says:

      “22kWh x 23.7M vehicles equals energy use of 521 GW hours/day.”

      This is wrong. Every vehicle does not discharge it’s battery 100% every day.
      You are a factor of 5-10 off there (unless you are trying to calculate some kind of theoretical maximum power limit).

      You need to calculate charging demand from daily mileage, not battery capacity.

      • Alex says:

        The article on 2050 electricity demand has a table using UK DfT mileages to calculate annual electricity demand.

        Vehicles will be charged when not in use, which normally coincides with lower demand for electricity – the possible exception to be managed being people coming home and wanting to charge after work.

  40. ralfellis says:

    Prof McKay, the government science officer, did this calculation many years ago. See his pdf booklet here.

    It is packed with data, and some total energy consumptions are on p103 – 109. I did not know that aviation was such a big sector. And even though Prof McKay was a greeney, he still concluded we cannot live on renewables. And that was even after he skated over the problem of intermittancy and storage.

    He also made the absurd claim that electric vehicles are 5x more efficient than fossil vehicles (having just said hydrogen vehicles were inefficient). This was picked up by the Sunday Times, who parroted this to the world as proof of how good electric cars are. And when I called out the Times and the prof, he threatened to sue me. After a length tussle, he made a grudging apology.

    The underlying conclusion of this informative booklet, is that renewable simply will not cut the ice, not matter how green-leaning you are. And nuclear seems to be the only viable option.


    • Wookey says:

      Electric vehicles are about 5 times as efficient as fossil vehicles (in fuel to motive effect terms). Maybe more like 4 times, but it’s not absurd.

      And in what respect was Prof McKay a ‘greeney’? Seems to me that he was a smart academic physicist with a good grasp of the energy system, and environmental issues associated with it.

      • ralfellis says:

        But you are leaving out the most important steps – the conversion of fossil fuels into electricity, and transmission to the socket. When these are added in, my diesel car is more efficient than any electric vehicle on the road.

        When calculating hydrogen car efficiency Prof McKay added in this step, but he negated it for electric vehicles. Hence the 5x claim was not simply wrong, it was pure disinformation. Socket electricity is still 80% fossil fuelled.


    • oldfossil says:

      Regarding SEWTHA. David MacKay was a cool guy. A strong advocate of FF replacement, he wasn’t afraid to name the downfalls of RE. He is reported as having said in a lecture that to carry the UK through a winter, a hydro storage facility the area of the Lake District and 500 foot deep would be necessary.

      In SEWTHA, he said that the energy density of petrol is 10kWh per litre.


      Compare the price per kWh of petrol and electricity.

      In the USA, electricity is about four times cheaper than gasoline. That makes EVs a no-brainer.

      In the EU, and let’s leave Germany and Denmark completely out of this shall we, we don’t want to die laughing do we, electricity is on a par with petrol. Your choice of engine will then shift to other factors such as purchase price, maintenance and resale value.

      If you’re a real petrol-head, driving a high-performance electric car with its almost linear power delivery must be the ultimate thrill. I remember my first Jap Crap motorbike that revved to ten and a half, and my RX-7 that only ran out of steam at eight and a half. I rode a Yamaha 250GP bike that only had two throttle settings, on and off. And a Tesla S handles as well. I would love to own one, but they are not available in my country, and my bank manager wouldn’t agree.

Leave a Reply