Sidebar<\/u>: As an example, per energy density data given in Source 2 below, lithium-ion batteries (an excellent and relatively high-energy-density current technology battery) top out at storing about 0.786 kWh per liter of volume. Doing the math, that means storage of the daily output of a single 400MWe power plant \u2013 which isn\u2019t a particularly large power plant – for future use would require a lithium-ion battery array forming a cube with sides a minimum of roughly 75 ft 7 in long<\/i>, and probably much longer when internal supporting structure and interconnections are accounted for. The reliability and maintainability of any such battery array would virtually certainly also be problematic. <\/p>\n
California\u2019s current electrical generation capacity is approximately 80,000 MWe \u2013 so such a hypothetical battery array could store at most roughly 1\/2000th, or 0.5%, of one day\u2019s worth of California\u2019s electric generation capacity.) You’d thus need 2000 of them to store one day’s worth of capacity – or allowing for those down for maintenance or otherwise offline, likely around 100 of them to store one hour of CA’s generating capacity<\/u>.<\/p>\n
Further: since batteries store DC and the power grid is AC, using them would also represent a net energy loss. Why? Because to use them, you’d have to convert AC to DC to charge, then reconvert DC to AC when you used the power. You’re looking at a substantial loss for each conversion – likely between 5% and 10% for each.<\/i><\/p><\/blockquote>\n
\nHowever, consumers don\u2019t consume \u2013 nor do utilities produce \u2013 a constant amount of power 24\/7\/365. Why? Because in the US, the daily load typically follows a \u201ctwin peaks\u201d curve \u2013 one peak during the 6AM to 10AM time frame, and a second during the 5PM to 10PM time frame. The midday load is typically lower than either peak (but still substantially higher than the daily low); that daily low occurs during the early morning hours (1AM to 5AM).
\n
\n
![](\"https:\/\/www.eia.gov\/todayinenergy\/images\/2020.04.06\/chart2.svg\")
<\/center>\n
\nThe minimum amount of load required 24\/7\/365 is defined by this early morning low. It\u2019s termed the grid\u2019s \u201cbaseload\u201d (Source 8), and is typically generated by large, fairly efficient power plants that operate continuously \u2013 and which typically take long periods of time to start up or shut down. Since many of them take hours to days to start up and shut down, these plants sometimes are forced to operate at part-load (or even idle) if demand is abnormally low. Such part-load operation is not efficient; these plants work best when working close to their design capacity. Examples include nuclear plants, coal-fired thermal plants, and many combined cycle plants (while the turbine portion of a combined-cycle plant can get up to speed quickly, depending on capacity and design the secondary cycle of a combined-cycle plant can take a while to get going from cold shutdown).<\/p>\n
For the rest of the day, the electrical grid must provide more power than the baseload to its customers. This additional power required over and above baseload is generally termed \u201cpeak load\u201d or “peaking power” (Source 9); it\u2019s typically provided by either smaller plants with relatively quick startup\/shutdown times, by renewable power sources (wind, hydro, or solar), or by transmission of power from areas with a local capacity excess to areas needing the power (which can result in substantial transmission losses and often much be purchased from other utilities). These plants tend to be either expensive in terms of operating costs (when fossil fueled, such as single-stage gas turbines or large diesel power plants) or sporadically and unpredictably unavailable (wind, solar). Even hydro\u2019s ability to meet peak demand can be limited by drought, min\/max flow requirements, or other environmental regulation. <\/p>\n
As you might have guessed, forecasting and satisfying peak load is a major challenge for utility companies. <\/p>\n
Finally, roughly 85% of generating plants are are based on some form of heat engine – and varying types have varying thermal efficiencies. (In simple terms, a generating plant\u2019s thermal efficiency is the fraction of the plant\u2019s heat energy input – typically from burning some form of fuel – that\u2019s converted into electrical energy.) Among fossil fuel plants, the Wikipedia article linked as Source 10 indicates that combined cycle plants are typically between 55% and 60% efficient. Source 10 also indicates that single stage plants such as single stage gas turbines, coal\/oil fired plants (if any of the latter are still in operation), and various types of biomass\/landfill gas\/waste incineration plants are usually around 37% efficient. (For reasons I discuss later, I\u2019m excluding nuclear or renewable generation other than biomass\/waste gas\/waste incineration.) Those efficiencies are under optimum conditions – typically at or near design capacity. Operate them under lower load and the efficiency generally goes down, sometimes dramatically.<\/p>\n
Data<\/u><\/b><\/p>\n
So, where can we get the data to answer the original question about the true fuel efficiency of electric vehicles? It turns out an electric vehicle blog provided that about 3 months ago. They did two range tests on the 2020 Chevy Bolt \u2013 an electric-only vehicle with a 66kWh battery pack. I\u2019m only going to discuss one of the tests; the numbers the two tests produced were within 1.8% of each other with respect to the amount of energy consumed by the Bolt per mile. You can read about the test at the link for Source 1 below if you like.<\/p>\n
Unfortunately, the article didn\u2019t attempt to actually calculate how much energy (in terms of E10 gasoline equivalent) was consumed to charge the vehicle\u2019s battery prior to test. However, the information in the article \u2013 along with other information readily available concerning power transmission losses, generation efficiency, electric vehicle charger efficiency, and how power plants are used by utilities – does allow us to determine that value.<\/p>\n
As the article indicates, the Bolt was driven 218.1 miles at a near-constant 70 MPH. It consumed 64 kWh of stored electrical energy while doing so. Simple math means that the Bolt consumed just over 0.293kWh per mile. <\/p>\n
So, what does that equate to in terms of E10 gasoline? Glad you asked.<\/p>\n
E10 Gasoline Equivalent Calculation Details<\/u><\/b><\/p>\n
It turns out that Wikipedia has a page concerning the energy density of various compounds, materials, and devices (Source 2). One of those compounds is E10 gasoline \u2013 e.g. gasoline with 10% ethanol by volume, which is the fuel most commonly used today in automobile engines. That page lists the energy content of E10 as 9.2167 kWh\/liter \u2013 which is roughly 34.839 kWh\/gallon.<\/p>\n
Holy schmoley! That means the Bolt used the energy equivalent of about 1.76 gallons of E10 gasoline! It got the equivalent of over 166MPG!<\/p>\n
Um, no it didn\u2019t. Because that\u2019s only part of the story. <\/p>\n
That\u2019s the amount of electrical energy provided from the Bolt’s battery while driving 218.1 miles. But that electrical energy didn\u2019t appear in the battery by magic. The above E10 gasoline equivalent figure does not<\/u> account for the energy used to produce that electricity; the energy lost while transporting that electricity to the vehicle; or for the energy lost while getting that electrical energy from the wall socket into the battery. Each of those steps impose a significant energy penalty.<\/p>\n
Let\u2019s look at charging the battery first. <\/p>\n
Auto batteries are DC; the power grid is AC. Thus AC-to-DC conversion is required to charge the battery from grid power. <\/p>\n
No AC-to-DC conversion is 100% efficient. Source 3 below indicates that in 2014, a typical electric vehicle battery charger was between 84% and 90% efficient. While charger technology has improved, I\u2019m pretty sure it hasn\u2019t improved hugely in the last 5 or 6 years (the 2020 Bolt’s charger would almost certainly have been designed no later than 2018). So to account for any improved technology let\u2019s assume that the Bolt\u2019s charger is 92% efficient vice between 84% to 90% efficient. That may well be too much of an improvement, but we’ll go with it anyway.<\/p>\n
Doing the math to correct for that loss, that means you need the equivalent of (1\/0.92) x 1.76 = 1.91+ gallons of E10 gasoline. And that\u2019s only the start of the corrections.<\/p>\n
Next, let\u2019s correct for transmission line losses. <\/p>\n
On average, per Source 4 approximately 5% of the power supplied by generating plants to the US electric grid is lost during transmission. That means we need to divide the E10 gasoline equivalent figure calculated above by 0.95 to correct for those losses \u2013 which in turn means that the Bolt now needs the energy equivalent of 2.014+ gallons of E10 gasoline. Perhaps not coincidentally, using this figure in calculating highway MPG equivalence would yield 108.27 MPG \u2013 which is almost exactly the EPA\u2019s \u201chighway MPGe\u201d of 108MPGe for the Bolt. <\/p>\n
However, even though the EPA seems to quit here, this isn\u2019t the true end of the story. And the biggie is yet to come.<\/p>\n
Outside of a thunderstorm, electricity doesn\u2019t just \u201chappen\u201d; it has to be generated. This generally is done by a process involving the production of heat \u2013 e.g., burning some form of fuel. (As I noted previously, I\u2019m excluding nuclear, hydro, wind, and solar. Later in the article I\u2019ll discuss why.) So now let\u2019s account for actually generating the electricity required.<\/p>\n
All forms of heat engine are less than 100% efficient; theoretically they cannot ever<\/u> be 100% efficient. And here, it gets a bit more complicated; we have three cases. Which case is appropriate depends on which type of power plant produced the electricity consumed by the electric vehicle.<\/p>\n
Case 1<\/b>: the electricity was produced by a single-stage generating plant. That means to find out how much fuel (in terms of the equivalent amount of E10 gasoline) was burned to produce the electricity in question, you need to account for the 37% typical efficiency of a single-stage generating plant. Or, in other words: divide the equivalent E10 gasoline of 2.014+ gallons previously calculated by 0.37 \u2013 yielding the energy equivalent of 5.444+ gallons of E10 gasoline. That’s how much heat energy that single-stage plant required, in terms of E10 gasoline equivalence, to produce that electrical energy. It got that heat energy by burning some form of hydrocarbon fuel.<\/p>\n
In this case, that means to go 218.1 miles at 70 MPH the Bolt “burned” \u2013 indirectly, at a remote generating plant vice on-board \u2013 fuel having the same energy content as a conventional vehicle getting just under 40.06 MPG would have used. Good? Yes, obviously. But it’s less than 40% of that \u201c108 MPG\u201d equivalent that the EPA claims for the vehicle.<\/p>\n
Case 2<\/b>: the electricity was produced by a combined cycle plant. Here, I’ll assume 58% as the plant\u2019s efficiency (just above the midpoint of the typical range for combined cycle plants). Using the 2.014+ equivalent fuel previously calculated that accounts for charger and transmission losses, that yields the need to use the energy equivalent of 3.473+ gallons of E10 gasoline to charge the Bolt. Or, in other words, generating that electricity required the combined cycle plant to burn fuel having the same amount of energy as a conventional or hybrid auto getting almost 62.8 MPG would have used. Yes, that’s excellent. But again: it’s also nowhere close to 108 MPG, either.<\/p>\n
Case 3<\/b>: we have no idea where the electricity came from. For this case, I’ll use the average efficiency for US fossil fuel generation plants. Sources 6 and 7 give the information necessary to calculate that \u2013 and it works out to just over 47.83% thermal efficiency (just under 51.59% of US fossil fuel plants now appear to be combined cycle plants; I assumed 58% efficiency for those and 37% efficiency for the others). In that case, burning the energy equivalent of just over 4.211 gallons of E10 gasoline is required. That in turn means the electric vehicle burned – indirectly at whatever combination of plants generated the electricity used to charge said electric vehicle – the same amount of fuel (in terms of energy content) as a conventional vehicle getting about 51.8 MPG. Again: that’s very good, but also nowhere even close to 108 MPG.<\/p>\n
Why No Nuclear, Solar, Wind, and Hydro<\/u><\/b><\/p>\n
Above, I didn\u2019t include nuclear or wind, solar, or hydro as sources. Why? That’s a fair question.<\/p>\n
There\u2019s a reason I didn’t include them. Or, more precisely, there are two reasons.<\/p>\n
Reason 1<\/b>: time of day considerations. This is why I excluded nuclear power. <\/p>\n
Nuclear power generates approximately 20% of US electricity. However, nuclear plants take literally days to start up and shut down. That means electricity generated using nuclear power is produced and used virtually exclusively to meet part of the US electric grid\u2019s baseload \u2013 e.g., that constant load present 24\/7\/365. <\/p>\n
This in turn means that every watt of nuclear power available is already “spoken fore” before the first electric vehicle is ever charged<\/i>. Electric vehicles will thus be charged using peak load generation assets. (Remember: any additional electricity used over and above baseload must perforce be generated by sources other than those providing the baseload.)<\/p>\n
That\u2019s significant for another reason, too. Human nature says that the overwhelming majority people aren\u2019t going to get out of bed at 1 AM each and every day to go plug in their electric vehicle every day (or, alternatively, rig a timer to ensure it doesn\u2019t start charging until then – assuming they can even find a timer that will handle the required load and\/or one that handles 240VAC). Rather, they\u2019re going to charge their electric vehicle either during the day at work (if they have a charging station available there) – or they’ll plug it in when they get home, probably between 6 PM and 8PM. <\/p>\n
Assuming only a 2 to 4 hour partial charge is required daily to \u201ctop off\u201d the battery (a full charge for the Bolt takes about 9 hrs), that means the vast majority of electric vehicle owners will be doing virtually all of their charging during peak load hours. This in turns means they’ll be using electricity generated by sources other than nuclear power plants – and won’t be increasing the grid’s baseload. <\/p>\n
Bottom line: unless someone charges their electric vehicle exclusively between the hours of 1AM and 5AM, that means they aren\u2019t using any<\/u> nuclear power to charge it for two different reasons. They’re using peak load power, not baseload power. <\/p>\n
Reason 2<\/b>: inherent limitations of wind, solar, and hydro. Even though wind, solar, and (to some extent) hydro are all generated primarily during peak load hours (1AM to 6AM is generally the calmest part of the day \u2013 and it\u2019s generally pretty dark then, too), as California is finding out the hard way these types of power are simply not always available when needed. Even under optimal conditions, solar isn\u2019t available for much of the evening portion of peak load and can also be substantially degraded by weather conditions. Wind power is similarly dependent on the weather. Even hydro is often limited by minimum\/maximum water discharge requirements, droughts, required reservoir levels, or other environmental considerations. <\/p>\n
So even if you have these sources available, you still have to have the capability to generate enough power by quick reacting conventional means to satisfy maximum demand – which means replacing their output if required. Otherwise you risk having power shortages and rolling blackouts. <\/p>\n
Why? Because other than hydro, you simply can\u2019t count on these sources always being available when needed<\/i>. And even hydro may be unavailable (or only partially available) due to regulatory or environmental restrictions.<\/p>\n
Since those sources may or may not be available at any given time (and some are known to be unavailable during part of peak load hours), accounting for them grossly complicates the analysis. I thus didn\u2019t include them in the analysis above. Collectively, they (solar\/wind\/hydro) account for about 15% of US total electric generation capacity. If someone wants to re-do my analysis above using the energy from that combination of sources while also using appropriate availability figures and other data to account for the known and average times those three sources aren\u2019t available, be my guest. That would change the numbers some in an electric vehicle\u2019s favor. But it won\u2019t change them radically; my guess would be somewhere around 10% when known unavailability periods and historical availability data are taken into account.<\/p>\n
Finally, before someone asks: no, I\u2019m NOT comparing apples and oranges here. Rather, the EPA (and electric vehicle vendors) are the ones doing that. And given the EPA\u2019s obsession with \u201call things green\u201d, it wouldn\u2019t surprise me to find out that\u2019s being done intentionally.<\/p>\n
The EPA appears to be making their comparison starting at point of consumer purchase \u2013 e.g., considering the purchase of electricity equivalent to the purchase of gasoline. What that ignores is the fact that in general, producing electricity means using some type of fuel – and if you’re talking other than baseload hours, between 80% and 90% of the time that means burning carbon-based fuel. Their method of calculating an \u201ceffective MPG\u201d for electric vehicles appears to ignore that fact entirely, treating the electricity used to charge an electric vehicle’s battery as if it were produced by pixie dust or unicorn smiles. <\/p>\n
A more honest comparison \u2013 and one that is a true apples-to-apples comparison \u2013 is to compare the energy content of the fuel used to generate the power needed for two different vehicles to travel the same distance (per mile, for example), regardless of where that fuel is consumed<\/u>. “Plug-in hybrids” excepted, a conventional or hybrid vehicle generates that power on-board with its engine; an electric vehicle depends on some external remote generator to charge its batteries or it goes nowhere. But in either case, in general today some type of fuel is burned to generate that energy. (And spare me the “fuel transportation losses” argument, please. Just like gas stations, power plants must have their fuel delivered to them too. That’s a wash.)<\/p>\n
A Real-World Comparison<\/u><\/b><\/p>\n
Electric vehicles are indeed energy efficient; the equivalent of between 40 to 63 MPG at 70 MPH for 3 hours is nothing to sneeze at. If you never drive anywhere that’s more than 90 to 100 miles away from your home, an electric vehicle like the Chevy Bolt might even serve your needs fairly well \u2013 provided you don\u2019t have to go anywhere for 9 hours after you get back, of course. If that\u2019s your \u201cthing\u201d, go ahead and buy one. Just don\u2019t ask me<\/u> to help subsidize your<\/u> choice to buy an electric vehicle with my tax dollars.<\/p>\n
But don\u2019t buy one thinking that you\u2019re single-handedly doing something major to \u201csave the planet\u201d. Generally, driving that electric vehicle will still require burning a fair amount of carbon-based fuel; that combustion will simply take place at a power plant vice at your vehicle. And it’s also additional fuel that the power plant wouldn’t otherwise have burned if you hadn’t used it. Remember, electricity doesn’t appear by magic; pixies and unicorns don’t exist. <\/p>\n
When things are calculated honestly, you haven\u2019t really reduced the amount of fuel burned to support your driving all that much, if any. All you did was move the source of combustion to a remote location.<\/p>\n
What you\u2019ve done is buy a fuel-efficient vehicle \u2013 but in reality it\u2019s nowhere near the \u201c100+ MPG equivalent\u201d vehicle that the EPA misleadingly claims it to be. When you actually do the math and account for the fuel used to generate the electricity needed to charge the battery, for the Chevy Bolt it\u2019s more like 40 to 63 MPG. And if you’re charging that Bolt anytime except between about 1AM and 5AM, it’s probably far closer to 40 MPG than 60 MPG. Why? Because in charging it, you’re using incremental power generated during peak load hours – and since none of that incremental power is nuclear, there’s a very good chance that the power you’re using to charge your electric vehicle is being generated by a quick-response single-stage power plant with 37% efficiency.<\/p>\n
In fact, an electric vehicle in reality is not much if any more energy efficient than some vehicles costing $10k less that don\u2019t have an electric vehicle’s drawbacks. And depending on the actual source of the electricity used to charge it, it can actually be less fuel efficient.<\/p>\n
Here’s a real-world example: within the past year, someone I know well purchased a new hybrid vehicle (no, not the \u201cplug-in\u201d variety \u2013 just a normal hybrid). That vehicle’s owner has taken it on at least three 1,000+ mile trips. <\/p>\n
At a fairly constant 70 MPH on the highway on reasonably level roads, that hybrid vehicle gets about 41 MPG. (The Bolt test referenced above was done mostly on the NJ Turnpike which as I recall is also fairly level, so that’s a fair comparison.) At a constant 60 MPH on reasonably level highways, it gets a bit over 47 MPG (1,200+ mile trip average). <\/p>\n
I\u2019ve personally checked those numbers by doing the MPG calculations myself manually using a calculator from fuel consumed and miles driven data recorded by the owner during the trips. My hand calculations were very close to the MPG info reported by the vehicle\u2019s onboard computer; any deviation was likely due to the presence or absence of a smallish “air pocket” in the fuel tank after the end-of-trip fueling.<\/p>\n
That vehicle has a one-way range of between 500 and 600+ miles on a full tank of gasoline (about 500 at 70MPH, and 600+ at 60MPH). Plus, getting enough fuel to drive another 250 miles – or another 500 miles, for that matter – takes 10 minutes or less vice 9 hours.<\/p>\n
And here\u2019s the best part: that vehicle cost about $10k less<\/u> than the list price of the base model<\/i> 2020 Chevy Bolt EV. (smile)<\/p>\n
—–<\/p>\n
Sources<\/u>: Source for image used in article: https:\/\/www.eia.gov\/todayinenergy\/detail.php?id=43295<\/em><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"
\n1 \u2013 Chevy Bolt Range Test: https:\/\/insideevs.com\/reviews\/423144\/chevy-bolt-ev-70-mph-range-test\/<\/em><\/a>
\n2 \u2013 Wikipedia Article on Energy Density: https:\/\/en.wikipedia.org\/wiki\/Energy_density<\/em><\/a>
\n3 \u2013 EV Battery Charger Efficiency: https:\/\/ieeexplore.ieee.org\/document\/7046253<\/em><\/a>
\n4 \u2013 Average Power Transmission Loss Data: https:\/\/www.eia.gov\/tools\/faqs\/faq.php?id=105&t=3<\/em><\/a>
\n5 – Car & Driver Chevy Bolt Review: https:\/\/www.caranddriver.com\/chevrolet\/bolt-ev<\/em><\/a>
\n6 – Fraction of Natural Gas Electric Generating Plants that are Combined Cycle: https:\/\/www.eia.gov\/todayinenergy\/detail.php?id=39012<\/em><\/a>
\n7 – US Electricity by Source Data: https:\/\/www.eia.gov\/tools\/faqs\/faq.php?id=427&t=3<\/em><\/a>
\n8 – Baseload Explained: https:\/\/energyeducation.ca\/encyclopedia\/Baseload_power<\/em><\/a>
\n9 – Peaking Power Explained: https:\/\/energyeducation.ca\/encyclopedia\/Peaking_power<\/em><\/a>
\n10 – Wikipedia article on Fossil Fuel Power Plants: https:\/\/en.wikipedia.org\/wiki\/Fossil_fuel_power_station<\/em><\/a><\/p>\n