TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Thoughts on Producing and Delivering Enough Energy to "Go Electric"

Discussion in 'Energy, Environment, and Policy' started by gizmoboy, Oct 1, 2015.

  1. gizmoboy

    gizmoboy Member

    Joined:
    Jul 2, 2015
    Messages:
    611
    Location:
    Torrance, CA
    I was wondering if anyone else had read Jean-Louis Gassee's MondayNote post about the VW scandal and the challenges he sees in generating and delivering enough power for a meaningful move to EVs.

    He's pessimistic about the ability of electric to match the energy density of petrol, and I'm curious if those who know the science better than I can see the flaws in his argument. If it holds up, I'd be pessimistic about a large-scale move to EVs as well.
     
  2. jdw

    jdw Member

    Joined:
    Jun 1, 2015
    Messages:
    169
    Location:
    Montreal, Canada
    The first thing that jumps out at me is that there is no adjustment for efficiency. A gas vehicle uses somewhere around 20% of the energy in a tank of gas to actually move the vehicle - the rest is used to produce heat and noise or is wasted in idle time. An electric vehicle is closer to 60-80% efficiency.

    You can get an idea of this from the "equivalent MPG" stat in the Remote S app which is usually around 120-130 "MPG".

    Alternatively, I use 18kWh to drive 60 miles, which would take 3 gallons of gasoline in a typical ICE vehicle. 18kWh is 64800000 Joules; 3 gallons of gas is 395280000 joules or over 6 times the total energy usage. Adjusting for city driving, electric vehicles are probably one tenth of the total energy usage of an ICE vehicle.
     
  3. nwdiver

    nwdiver Active Member

    Joined:
    Feb 17, 2013
    Messages:
    2,393
    Location:
    United States
    #3 nwdiver, Oct 1, 2015
    Last edited by a moderator: Mar 13, 2016
    WOW... there are so many things wrong with that hit-piece it's hard to know where to begin.

    - Why convert gasoline in to Joules? kWh are far easier... Ah... with joules you have to use scary scientific notation...
    - As mentioned above... EVs use >60% less energy than ICE.
    - <10% of charging is done under the 'gas station' paradigm; The other 90% is done at home @ <20kW.
    - Even for an ICE... 20gallons per week is not the average US consumption...

    I couldn't help but think of this scene while I was struggling to get to the end of the article...

     
  4. jdw

    jdw Member

    Joined:
    Jun 1, 2015
    Messages:
    169
    Location:
    Montreal, Canada
    #4 jdw, Oct 1, 2015
    Last edited: Oct 1, 2015
    Second thing that jumped out at me was the assumption that gasoline is a naturally occurring resource. It takes a lot of inefficient energy consumption to find, extract, transport, refine and transport gasoline to the pump. If that energy was used to produce electricity at the efficiency level of a typical power plant, it would probably be sufficient in itself to power a very large number of electric vehicles.
     
  5. MichFin

    MichFin Member

    Joined:
    May 8, 2015
    Messages:
    281
    Location:
    Detroit, MI
    Almost all electric car charging happens at night when electricity production is usually idle. That's why I get 1/2 price electric rates with my off peak charging. I read a study 5 years ago that if all cars in Israel would switch electric it would only require an 8% increase in grid capacity.
     
  6. nwdiver

    nwdiver Active Member

    Joined:
    Feb 17, 2013
    Messages:
    2,393
    Location:
    United States
    #6 nwdiver, Oct 1, 2015
    Last edited: Oct 1, 2015
    So here are the salient 'elevator points'...

    - The US currently uses 270B gallons of petroleum products per year.
    - For simplicity converting that to electric reduces that energy by 2/3 to ~90B gallons equivalent.
    - 90B x 35kWh = 3150 TWh/yr
    - Current US electrical consumption is ~4000TWh/yr

    How much area to generate 7500TWh/yr from solar alone?

    - A 10MW solar farm can generate 25GWh/yr and occupies 0.04 square miles.
    - You would need 300k 10MW farms occupying 12k square miles.
    - For scale there are 3500 square miles of golf courses in the US and 61k square miles of paved surface.

    The numbers aren't nearly as overwhelming when you don't use Joules..... probably why he used Joules :wink:
     
  7. AudubonB

    AudubonB Mild-mannered Moderator Lord Vetinari*

    Joined:
    Mar 24, 2013
    Messages:
    4,219
    Location:
    Denali Highway, Alaska
    Thanks for the elevatadata ;)

    And now to make them even more crunchable and usable: how many square miles in parking lots.....and why cannot most of them be dual-purposed, with solar panels above them?
     
  8. nwdiver

    nwdiver Active Member

    Joined:
    Feb 17, 2013
    Messages:
    2,393
    Location:
    United States
    #8 nwdiver, Oct 1, 2015
    Last edited: Oct 1, 2015
    The number of parking spaces is actually harder to nail down than you would think... estimates range from ~800M - 3B. You would need to cover ~3B parking spaces with solar PV so that might work...

    IMO the 'pithiest' retort would be 'We need ~3x as much space for Solar as we use for golf' and yes, mostly on roof tops and parking canopies.
     
  9. ecarfan

    ecarfan Well-Known Member

    Joined:
    Sep 21, 2013
    Messages:
    10,316
    Location:
    San Mateo, CA
    And as Elon pointed out in the Tesla Energy launch presentation earlier this year, most of the solar energy power generation we need to add to move the grid onto renewable energy can be added to existing building roofs, very little unused land is needed.
     
  10. cpa

    cpa Member

    Joined:
    May 17, 2014
    Messages:
    933
    Location:
    Central Valley
    Diver,

    would we not be able to redirect some of the electricity used to refine, extract, etc. etc. petroleum towards charging all the BEVs out there? It would seem to me that amount would not be an insignificant amount.

    It was not clear whether your calculations factored this in or not. Regardless, certainly not as overwhelming as implied in the story.
     
  11. nwdiver

    nwdiver Active Member

    Joined:
    Feb 17, 2013
    Messages:
    2,393
    Location:
    United States
    #11 nwdiver, Oct 1, 2015
    Last edited: Oct 1, 2015
    Actually a relatively small amount of electricity is used for refining... <300w/gal if I recall correctly. ~7kWh/gal ENERGY is used but most of that is Nat gas.


    LOL That's an understatement... Quote from the article;
    [FONT=monday_text_egyptian_webRg]
    'Even if we dedicated the entire Earth’s surface to solar energy collection and conversion, either photovoltaic or photosynthetic, electronics or plants, we couldn’t possibly match our consumption.'[/FONT]
     
  12. snort

    snort Member

    Joined:
    Jun 27, 2015
    Messages:
    105
    Location:
    Seattle, WA
    He's not trying to factor in local generation, such as rooftop solar. Using a fairly bad-case scenario: a Seattle based 20kw nameplate system generates an average of about 20kwh per day in the winter. (About 80kwh in the summer.) This takes less than 1400 square feet: e.g. 40*35, a fairly typical small American suburban rooftop. 20kwh can move a model S about 60 miles; the Leaf is better and we can assume that will be true of the model III as well. That's all most people need most days. No impact to the grid AT ALL.

    of course this is an average--a long stream of dreary days will leave you wanting. adding a battery to the system (whether it's a local system like the PowerWall or net metering on the grid, pretending that the power companies do the right thing) costs about 25% efficiency loss. If you can only charge your car at night, you'll need to use something like this. But it still gets the job done. local batteries don't impact the grid at all (in fact they'll probably reduce the load). Grid attached storage, co-located with substations, or anywhere that's convenient, can achieve something similar. The power companies should do this anyway.

    Workplaces and highrise apartments are more of a problem. A typical tall apartment building has less than 40,000 square feet. fewer than 30 cars at this level. Apartment dwellers tend to use more transit than suburbanites, so this may be good enough. A car parked under a 10x20 solar panel on a Seattle winter day adds less than 3kw--less than 10 miles--if they're under there the whole time the sun is up. This is borderline adequate during the summer, not in the winter. But, the factory rooftop can also be used. If the factory produces more than it needs, it can charge cars. Electricity is fungible. Lots of factories do their own generation--solar, wind, etc., already. No reason to imagine that that won't expand.

    (an interesting bit of history: In the early days of the power grid, many of the big factories, especially steel mills, which produce a lot of excess heat, added generators to capture this and sell it. It cost them next to nothing to generate, so they wound up undercutting the electric utilities...who fought back by convincing most legislatures to ban the practice. Most of these laws, preventing what is today called co-generation, are still on the books. Even if the steel mills were still in the US, which most are not, it would only contribute a few percent to the grid if it were legal. Moreover, modern mills produce a lot less waste heat, even if they haven't been sold to the far east for short term profits by their owners.)

    Gassee's conspicuous omission of Tesla's gigafactory and the expansion implied by that tells us that he's either more ignorant than his history would suggest, or he's got an anti-ev agenda for some reason.

    --Snortybartfast
     
  13. jerry33

    jerry33 S85 - VIN:P05130 - 3/2/13

    Joined:
    Mar 8, 2012
    Messages:
    12,743
    Location:
    Texas
    The fly in this ointment is that almost all roofs are pitched, so only half or less of the 1400 sq.ft. is available for solar panels. I'd love to have them on my roof (1800 sq.ft house), but so far the best they can do is about 40% of my energy use. A high month (means A/C at 21C 30-40C outside, plus two EVs) is $170. Average is more like $110. I doubt you can get a system for $50 a month, and I'd still have an electric bill. So I use the 100% renewable energy plan.
     
  14. nwdiver

    nwdiver Active Member

    Joined:
    Feb 17, 2013
    Messages:
    2,393
    Location:
    United States
    Where in TX are you located Jerry? DIY systems aren't too difficult and the cost is now ~$1.2w ($0.84/w after FTC). If you shopped around a bit you might even be able to build a nice 8kW system for ~$8k ($5600 after FTC). 8kW in TX depending on location would generate ~14000kWh/yr.
     
  15. snort

    snort Member

    Joined:
    Jun 27, 2015
    Messages:
    105
    Location:
    Seattle, WA
    900 sqft can give you about 13kw nameplate. In Dallas, this can do about 68 kwh/day in July, 39 in December. Ample for your car, which was the subject of the essay. The good news is that solar is at its best exactly when AC is most needed. electricity is fungible. Anything you do to reduce the burden on the grid is to the good.

    Today, a lot of roofs are not pitched optimally for solar. we should not imagine that the way things are is the only way things can ever be, as Gassee is doing. Houses don't last forever and new ones are being built every day. Eventually, most will change.
    --Snorty
     
  16. Saghost

    Saghost Active Member

    Joined:
    Oct 9, 2013
    Messages:
    2,933
    Location:
    Delaware
    #16 Saghost, Oct 1, 2015
    Last edited: Oct 1, 2015
    As a couple of folks have mentioned, the analysis he did is fundamentally flawed, being based on replicating the energy content in the gas tank and some random assumptions about how often people refuel.

    I actually took my own stab at this problem some months back, based on the best metrics I could assemble for cars on the road and miles driven as well as statistics for the US power grid, with links for where I got the data included. I've been over it a couple times, but I don't guarantee I don't have a mistake in here somewhere, or an odd assumption.

    I based it on the assumption that every car in the US was suddenly replaced with a RWD Model S...
    Walter

     
  17. Theshadows

    Theshadows Active Member

    Joined:
    Apr 20, 2013
    Messages:
    1,896
    Location:
    PA
    For a very very conservative figure. Each 1kw of PV installed will get you 3000 miles in a Model S per year.

    This is at 40 deg north latitude. 50 deg north you would need about 1.1kw, 30 deg, 0.9kw.
     
  18. bluetinc

    bluetinc Member

    Joined:
    May 11, 2009
    Messages:
    682
    Location:
    MD
    #18 bluetinc, Oct 1, 2015
    Last edited: Oct 2, 2015
    Gizmoboy,

    That's quite a convoluted comparison that he generated. Here is a more strait forward way to look into the future. Lets look at a pathological case and say we could change all passenger miles driven in the US to EV miles today. Feel free to change the number to ones you like, I'm using easy numbers to calculate, and discounting charging and transmission losses.

    Average energy per mile .250kWh
    Average miles driven per year in the US 14,000
    Average number of cars on the road in the US 200,000,000

    That means each car will use (in one year) .250 kWh/mi * 14,000mi/yr = 3,500 kWh/yr
    And for the year, for the US: 3,500 kWh/yr * 200,000,000 cars = 700,000,000,000 kWh/yr

    About 4,000,000,000,000 kWh/yr of electricity were produced in 2014 in the US

    That means that you (might, worst case) need to produce 4,700,000,000,000 kWh of electricity over the next year, or about 20% more than we did last year. That's not bad, it's a pathological case that doesn't take into account that it will take us 20 years to even attempt to get there, that most charging will happen in off hours which are easier for the utilities to deal with, or that the change over to solar that is picking up.

    Oh, and just a quick thought on solar since we are looking at these numbers. That energy you need to drive for a year (3,500 kWh/yr) can be produced where I am (MD) with a 2 kW solar system, small by today's standards. That system takes up about 10x10 rooftop area (100 square feet), which is about the area of roof just above where the car is parked.

    Peter



    Used these site for rough numbers:
    How Many Miles Do Americans Drive Per Year?
    What is U.S. electricity generation by energy source? - FAQ - U.S. Energy Information Administration (EIA)
     
  19. jerry33

    jerry33 S85 - VIN:P05130 - 3/2/13

    Joined:
    Mar 8, 2012
    Messages:
    12,743
    Location:
    Texas
    DFW.

    However, I'm a firm believer in not giving a screwdriver to a software guy (or a code patch to a hardware guy), plus I'm not good on ladders. No way would I be able to convince myself that I could successfully do a DIY solar installation. The last quote I received was in the $30K area a few months ago. Solar City came up with an excuse that they wouldn't do any roof that was steeper than 23 degrees (mine is 25).
     
  20. jerry33

    jerry33 S85 - VIN:P05130 - 3/2/13

    Joined:
    Mar 8, 2012
    Messages:
    12,743
    Location:
    Texas
    The 250 is about right for me (248 actual), but I'd suggest a better average is 350 kWh/mi for the majority of Tesla drivers (The Model 3 may bring this down to closer to 250, but that's speculation at this point).

    I'd suggest that the average Tesla driver puts on more than 14,000 miles a year. My average has been 1850/month 22,200/year. As EVs get more range, people will likely switch from planes to EVs for the majority of vacation trips (that don't involve crossing a large body of water). I suggest this will happen because between the airline deregulation and all the security checks, air travel has become a worse experience than bus travel--unless of course you fly your own plane, then it's a far better experience.

    Actual energy used is 14.3 MW for 57,600 miles. This doesn't include metre to charger losses or pre-warming.

    About 50% of the driving is has been for trips.

    So I think the target generation should be somewhat higher than your projection (the main premise of "it will be a long time before the extra generation is needed" is correct).
     

Share This Page