Carnegie Mellon University Studies Electric Transportation
Glenn Doty points out a flaw in my recent piece about electric transportation. He writes:
(The Carnegie Mellon University study) assumes a life-cycle grid emissions profile of 615 g-CO2E/kWh. That is blatant BS.
The impact of new marginal electricity demand (as represented by shifting transportation demand from liquid fuel to electricity) can only be satisfied by spare generating capacity. There is no renewable spare capacity in most of the country, and in the places where there is spare capacity (TX, IA, MN, ND, IL…) there is no benefit to be had from a constant 8+ hour nighttime demand increase, as the spare renewable capacity in these cases is curtailed wind, and the constant 8+ hour night-time demand would be satisfied by not tamping down baseload power as much each night.
So this new demand must be assumed to be met with 100% fossil energy – not “average grid energy”. The average CO2 emissions from current fossil sources on the grid is ~875 g-CO2/kWh… that’s before upstream emissions are considered, and before methane emissions from wellhead and distribution leaks are considered. A more reasonable estimate should be ~1000 g-CO2E/kWh.
I reply:
Glenn:
You make a good point here, as always, but I urge you to consider the bigger picture. There are several different important variables here, each changing in radical but uncertain ways as we go through the long process of replacing 230 million cars and trucks in the U.S. (over one billion world-wide). Please consider this:
1) We’re running out of cheap, abundant oil. No one knows the exact rate, nor any of the other data with any precision, e.g., the ecological and financial cost of extracting and refining increasingly gooey and toxic tar sands.
2) Even if we weren’t running out, oil would still be killing us in many different ways, none the least of which is the wars we prosecute to maintain access to it.
3) All auto OEMs and dozens of start-ups are sending us in the direction of electric transportation, the cost of which is coming down dramatically. E.g., check this out: http://2greenenergy.com/big-advancements-batteries/20808/.
4) Thus the value proposition for the consumer is improving every day, AND
5) The cost of renewable energy is approaching grid parity.
In my mind, all five of these points conspire to make electric transportation something of a slam-dunk over the coming couple of decades.
Renewable energy is good. Electric cars are good. Bigger and/or more centralized electric power plants are bad. Electric vehicles cannot be touted as a solution as long as the hidden costs of generating and transmitting the electricity over long distances is hidden and/or subsidized.
The solution is a large number of stand-alone, small-scale, solar thermal energy plants, of which the Solar Furnace CHP is one variety: http://solarfurnaceCHP.wetpaint.com. Another is Sopogy.com.
The focus on large, centralized electric generating plants seems fixed in everyone’s minds, as if no one has taken to time to read about small-scale locavore energy plants placed next to their respective loads. As in transportation issues, the main issue with load balancing and distribution is the “last mile”. That problem is solved with the small-scale, locavore, stand-alone solar thermal energy plant using dual axis Sun Trackers.
Jim Miller
For all the reasons that Craig listed, fossil fuels must be phased out. But with renewables? Approaching grid parity? I suppose that if the cost of renewables were 1000 times higher than grid parity (it isn’t) but were decreasing, it could be said to be APPROACHING grid parity. And, the grid would have to be greatly modified and expanded to enable it to collect and transmit power from widely scattered and diffuse sources; that would greatly add to costs.
Lord Kelvin, after whom the Kelvin temperature scale was named, stated that knowledge that cannot be expressed in numbers is not knowledge; that it may be the beginning of knowledge, but it is of a meagre and unsatisfactory sort. We need numbers arrived at by clear and objective methods, even if those numbers cannot be exact.
Most renewable sources of energy are intermittent. That has to be taken into account when comparing costs. For example, if a group of wind farms cost the same as traditional methods of generating power, to compare the cost with traditional methods, the cost of the wind farms would need to be multiplied by a factor to compensate for the fact that they are intermittent sources of power. The same is true with solar. Then, the costs would not be anywhere near comparable.
It has been argued that interconnecting wind and solar sources would provide reliable power. However, that hypothesis has never been tested; it is simply assumed to be true. Before commitments are made to an energy technology, its practicality should be ascertained by very thorough studies; again, that has not been done. A thorough study could be done if sensors were placed in all the locations where it would be reasonable to have wind and solar power generation systems. The data would, in real-time, be transmitted to a central location for real-time analysis to determine whether there are any periods when there would not be adequate power available.
I have search in vain for studies that actually verify that wind and solar power could do the job. It appears that no such studies exist. So, we are being asked to spend untold billions of dollars on technologies the practicality of which has never been verified.
It’s interesting that Deutchland, which has committed itself to PV power, has the second highest cost of electricity in Western Europe. Holland, which has committed itself to wind power, has the highest cost of electricity in Western Europe. France, which generates almost 80% of its power nuclearly, has the lowest cost, although France’s figures could be a bit iffy since its generating stations are government owned and I’m not sure whether there are any subsidies.
Consider also other countries. India is a very poor country with a population density several times greater than ours. It will take cheap energy to lift the poor people of India out of poverty. If they were forced to depend on such sources as wind and solar, they would never escape poverty. And, with the high population density, the land area for diffuse sources of power is very insufficient. Even now, their lives are shortened by the pollution resulting from burning wood for cooking and kerosene for lighting.
In 20 years, France went from 0% of power generated by nuclear reactors to 80% of its power generated by nuclear reactors. That would seem to undermine the argument that shifting to nuclear is too slow.
I’m not suggesting copying what France did; I think that probably they (and the rest of the world also) chose the wrong nuclear technology and that LFTR nuclear technology should have been developed and used because its safer, more economical, and solves the nuclear waste problem. If for some reason the LFTR doesn’t work out, there still are better nuclear technologies than what we have chosen.
Craig,
I still don’t think that you see the true impact of electrifying our transportation. Again, it still – and always – comes down to spare capacity.
As renewables grow and penetrate more deeply into the grid, they are fully utilized. So the amount of spare capacity of renewable energy doesn’t change, even as the amount of renewable energy continues to increase.
To illustrate, consider this: in 2010, there was ~167 TWh of “other renewables” (wind, solar, geothermal, and biomass firing) of electricity generated, or ~4% of the grid. In 2011, there was ~195 TWh of “other renewables”. That’s a 28 TWh, or ~17% increase in renewable energy!!! That is HUGE, and EXCITING, and I am not in any way attempting to belittle that. It’s a BIG deal.
But the amount of SPARE renewable capacity (blanketed solar panels, tamped biomass firing plants, powered-down pumps for geothermal energy plants, and curtailed wind) equaled ~3 TWh (all curtailed wind) in 2010, and likely equaled less than 3 TWh (all curtailed wind) in 2011… and will likely equal less than 3 TWh in 2012 (again, all curtailed wind – which EV’s can’t utilize).
Meanwhile, the growth in spare coal capacity has increased significantly – as more coal power is supplanted by cheap gas power and renewable energy.
So when you are looking at tapping exclusively spare capacity, you are going to be tapping more and more coal as renewables build out, because renewable energy is not building out any spare capacity.
Another way of saying this: When you put up a solar farm, you produce electricity, which causes less coal to be burned to produce electricity… When you plug in your car, you require more new energy to be produced, which means they have to power the coal plant back up to produce more.
That will remain a constant truth until renewable energy supplants almost all fossil energy. If it continues at the shocking rate of 17%/year (which is almost unthinkable), it will take over 30 years before we can actually start using renewable energy when we plug our cars into the grid… That’s 3 times the longevity of a battery pack that is purchased to power a car today.
An electric car purchased today will likely not see one miliwatt of consumed renewable energy… It will all be coal and natural gas, and each year the proportion of coal will increase for at least the next decade – the lifetime of a current EV.
Glenn:
Thanks for this. John Addison writes in EVWorld:
“My Nissan Leaf is powered by my utility PG&E with a typical California energy mix of 47% natural gas, 20% nuclear, 16% large hydro, and 15% other renewables. Yes, during peak summer afternoon demand, PG&E does import 2% coal power from other states, but I charge my electric car off-peak after 10 p.m.”
Isn’t it true that he’s using no coal?
Craig,
In truth, he’s not technically using coal. He is, however, likely causing more coal to be consumed in a ratio equal to the amount of nighttime energy he is using to charge his EV.
PG&E tamps back their natural gas to the minimum at night, using only fast-response peakers… but they produce plenty of nuclear energy and hydroelectric energy, and they import energy from CAISO…
Remember here that the electron impulses that we think of as electricity are indistinguishable from one another once produced… You buy energy, it is drawn off of the grid. So Addison claims that PG&E only imports 2% of their energy profile in coal because during summer peak they purchase some percentage of their energy from a provider that has a coal power plant… and then Addison does a straight calculation that states that only 2% of their profile is coal.
But the electron impulses are fungible: they are part of the grid. California imports ~25% of its electricity. Some of that energy that is imported is coal, other energy is hydroelectric, and some is wind, and some is natural gas, etc… But the snapshot grid mix at any given moment is still utilized… So any change in the dynamic of the grid (say – plugging in an EV), must then result in changing the grid mix by an similar amount.
At night, if PG&E had a policy that they must satisfy their power demand with their own generators unless they cannot do so, and then and only then would they purchase energy from CAISO, Mr Addison’s point would be a correct one: people who purchase energy from PG&E would be using only natural gas, not coal. But PG&E doesn’t have such a policy. I can say this with confidence because no participant in an ISO could enact such a policy without getting rioted by their consumers.
Instead, what likely happens is this: Mr Addison plugs in his EV, and PG&E either begins purchasing more energy from CAISO or begins selling less energy to CAISO (depending on where Mr Addison lives). CAISO, in either event, now has less energy, and must import more energy. At night, natural gas sourced energy is far too expensive to consider outside of demand response (peakers), so they’ll simply import more energy from Nevada and Oregon. Nevada and Oregon will respond by either reducing the amount they tamp down their baseload generation (we’re talking about an overnight 8+ hour recharge, it will certainly be met with baseload generation, not demand response), or they’ll sell less energy to Idaho, Colorado, Arizona, or they’ll import more energy from Washington…
However many states are staggered along, no-one is going to be burning any more natural gas than they have to in order to meet this demand… It’s too expensive. No-one is going to ramp up hydropower to meet the demand, because that leaves less hydropower to use during the peak demand period in the middle of the day – lessening the value of the energy stored behind the dam… and no-one can ramp up nuclear energy (it’s run full-out). So however long the trail runs, from seller to buyer, at the end of that chain is likely going to be a coal power plant.