Man Bites EV: Will EV Bite Back?

Some time ago, the Chevy Volt attracted my attention. I think the plug-in hybrid concept hits the sweet spot for American drivers, and the Volt’s 35–40 mile electric-only range seemed to be the perfect number. A pure electric vehicle (EV) would not permit my wife’s periodic work-related jaunt to Pasadena, so any battery-powered solution for us must be of the plug-in hybrid electric vehicle (PHEV) variety. The problem, ultimately, was the high price tag (and the hump in the middle of the back seat occupied by the battery). Although I don’t self-identify as being in the “upper class,” our income edges us into the top quintile in the U.S. So for us to decide that the Volt costs too much—despite genuine enthusiasm—seemed to spell trouble (indeed, the average income of Volt owners was claimed to be $175,000). My conclusion was that electric/plug-in cars are out of reach, and could well remain so.

In April of this year, I became aware of the Ford plug-in, called the C-Max Energi (yes, with an “i” at the end!). The C-Max Energi has a 21 mile electric-only range, and gets an EPA rating of 43 miles per gallon (2.3 gal/100 mi; or 5.4 L/100 km). The price tag is approximately $6k cheaper than the Volt, and the back seat passed my wife’s approval. Nonetheless, after carefully considering the C-Max Energi as a replacement for our increasingly ailing car, we decided against springing for one: still too expensive. I was all set to write a Do the Math post to the tune of “Almost bit on a PHEV again.”

But the fact remained that our 11-year old 28 MPG car (bought used) has been costing us a fair bit in maintenance, its reliability increasingly dubious. Replacement loomed. Motivated by an upcoming long-haul road trip, we explored options again, looking at hybrids and the C-Max Energi. In the end—aided by a federal tax credit, a California rebate, and an unfathomably good offer that together knocked $9k off the MSRP—we drove an Energi off the lot under battery power.

It turns out that:

• the lifetime cost for the PHEV is still higher than other options we considered, but not prohibitively so given credits, rebates, and discounts;
• the CO₂ emissions are cut in half in electric mode (considering upstream electricity production in our region);
• batteries still stink compared to liquid fuel, and likely always will.

 

EV Pro or Con? Decide, Dammit!

I remain skeptical that EVs or PHEVs will capture a large fraction of the U.S. market share. Yet I just voted with my own dollars to get one. Does this make me a hypocrite? A double-talking, contrarian, dirty hippy? Not in my view, naturally. I’m a very unusual consumer: hyper energy-conscious, generally frugal despite a relatively high-income, but all the same prone to take on energy-related hobbies that may not be a win in the strict financial sense. In this case, $3750 of federal tax credit, $1500 from California, and $4000 off the MSRP (corresponding to a shocking $2000 below invoice price) conspired to make the choice attractive and affordable. But these three discounts do not speak to the steady state fate of EV cars. The first two will expire at some point, and the deep price drop likely signals a panic from Ford responding to disappointing sales numbers that could portend doom and lost investment for the C-Max Energi line. These cars won’t be sold indefinitely at a loss. So I bought the car under highly unsustainable pretenses. Optimistically, maybe the incentives provide a necessary kick start while EVs become cheaper. Time will tell.

So far I am very pleased with the car: no cut corners, as far as I can judge. Around town, we basically have a pure electric car, but also have executed a very enjoyable 3,000 mile roadtrip. I’m swimming in interesting data, and appreciating transportation through new eyes.

But the fact that I now own a PHEV is not enough to transform me into an unabashed supporter, as often happens to early adopters. EVs are not the cheapest option: even just on the fuel front. With $4/gal gasoline and $0.15/kWh electricity, a half-and-half electricity/gasoline mix is neck and neck with a Toyota Prius getting 50 MPG. Add the initial battery cost and we more than wipe out the marginal savings in the cost of propulsion. I came under serious fire from Volt owners for once suggesting that the financial savings were a wash, or even—heaven forbid—negative. I get it: the emotional investment is large, and it’s hard to remain objective after spending $40,000. I will personally strive to steer clear of the attachment bias, and remain objective about the merits of electric cars.

I should also point out that making estimates of propulsion costs over the next 10–15 years is very difficult, because it is not clear whether gasoline costs increase or decrease over that timeline. Long term, they are almost certain to rise. But a spurt of fracking-produced oil—even if a limited-time offer—may hold prices down for a while. Meanwhile, slowly transitioning our electricity infrastructure to less carbon-intense forms, which I am all for, may drive electricity costs up.

Still, as I often find, applying a strict dollars-and-cents assessment imposes a terribly narrow window on the world. There are plenty of other reasons that I was attracted to a plug-in, even if it winds up costing me more money in the long term. Why do I have an off-grid photovoltaic system (with expensive, disappointing batteries)? Or a whole-house energy monitor? Or a 600 gallon (2300 L) rain catchment system? Or three chickens in the backyard? None of these choices are primarily financial in nature. The enjoyment I get out of quietly tooling around town, logging charge and mileage data like a madman, and developing the capability to self-charge off my own roof (even if the grid is down) offer recompense. Part hobby; part practical; part hedge against an uncertain future.

We’ll get back to some basic EV math in a bit. First, we’ll take a detour into environmental factors.

CO₂ Emissions

While climate change is not a primary motivator for me (resource depletion, growth reliance, and fossil-fuel dependence in general are my main concerns), I do take it seriously. If I’m unjustified in worrying about a resource crunch on a shorter timescale, and we therefore continue profligate consumption of fossil fuels, then climate change is there to make sure we get bitten either way.

So on that count, I am happy to report that driving the C-Max on electricity (in California) produces less than half the CO₂ that driving the same vehicle in hybrid (gasoline) mode. In fact, California analyzed different fuel sources for light-duty vehicles, finding that gasoline produces 96 units of CO₂ to the electricity mix value of 41.

If I use the window sticker values for the C-Max, driving 100 miles consumes 2.3 gallons of gas, or 34 kWh of electricity (from the wall outlet). Gasoline—including refinement costs—produces 12.6 kg of CO₂per gallon (96 g/MJ), while the California mix of electricity comes in at 0.446 kg/kWh of delivered energy (about a pound per kWh: numbers from here). So 100 miles of driving the C-Max on gasoline emits 2.3×12.6, or 29 kg of CO₂, while the electric option yields 34×0.446, or 15 kg.

Another triangulation comes from a handy EPA site that lets you determine your local electricity mix (by zip code), along with a figure for carbon intensity. The national average electricity is 44.5% coal, 23.3% natural gas, 20.2% nuclear, 6.8% hydro, 3.6% non-hydro renewables, and 1.1% oil. That amounts to about 70% from fossil fuels. For California, it’s 7.3% coal, 53% natural gas, 14.9% nuclear, 12.7% hydro, 10.1% non-hydro renewables, and 1.4% oil, totaling 61.7% fossil fuel (dominated by less carbon-intense natural gas).

The site puts the CO₂ intensity at 1216 pounds/MWh nationally, and 659 lb/MWh in California. I was also interested to see that despite a 46.5% hydroelectric contribution, Washington State has a CO₂ intensity of 819 lb/MWh: larger than California, owing to a 30% coal dependency.

At the EPA rating of 34 kWh/100 mi, 1 MWh would propel the C-Max Energi 2940 miles. The same car gets 43 MPG on gasoline, so that this trek would require 68 gallons of gas, producing 860 kg of CO₂ by our previous conversion, or 1900 pounds. This suggests the amount of CO₂ produced by gasoline is 2.9 times higher than by electricity in the same car. The disparity between the two estimates stems from the fact that the California government puts the CO₂ intensity of its electricity at 124 g/MJ, translating to 980 lb/MWh—50% higher than the EPA number.

In either case, it is clear that driving a car on electric propulsion can offer a net savings in CO₂ emissions—especially in California. Picking on my home state of Tennessee, obtaining 59% of its electricity from coal (and only 9% from hydro, despite the Tennessee Valley Authority system of dams) puts its carbon intensity at a little more than double that of California. In such places, it’s questionable whether electric drive produces a net CO₂ benefit. In places like Wyoming, Kansas, and Missouri, it is decidedly worse to tool around in an EV powered by utility electricity, from a carbon standpoint. The national average carbon intensity is 1.84 times the California value, according to the EPA. Here, too, the question of net benefit becomes mushy. One lesson is that it may be wiser to drive toward low-carbon sources before driving the country on electric cars.

Other Pollutants

While power plants do nothing to capture CO₂ emissions at present, they do tend to be proficient at scrubbing other pollutants, like nitrous oxides (NO_x) and sulfur dioxide (SO₂) from the exhaust stream. Catalytic converters in cars achieve some reductions (on NO_x), but we can’t expect a compact, lightweight, mobile device to perform as well as a giant piece of fixed infrastructure. The EPA site also presents intensities of these two pollutants stemming from regional electricity production (in graph above). Here, California shines again: for NO_x, the national average is 1.12 pounds per MWh, while San Diego gets 0.42 pounds per MWh. For SO₂, it’s even better: 0.18 lbs/MWh in California vs. a national average of 3.08.

On the flip side, the manufacture of EVs and PHEVs incur greater energy costs than do conventional cars, and also employ rare earth elements in the motors and involve caustic chemicals in battery production. A recent article in IEEE Spectrum surveys studies that put the net environmental impact of EVs slightly worse than that of conventional cars—despite achieving CO₂ reductions in propelling the vehicles. I have not personally delved into the numbers and analysis, but the result is credible. Assuming the conclusion applies to the national average electricity mix, the fact that California undercuts the national average CO₂ emissions by nearly a factor of two (and even better on other pollutants) means that EVs in California are very likely still a net environmental win—although not dramatically so. This again illustrates the importance of switching our electricity supply before (or at least in tandem with) large scale adoption of electric transportation.

Batteries Stink

I have warned before that electric vehicles are not obviously going to provide a viable large-scale path away from fossil fuels. In a connected vein, I have also expressed disappointment in batteries in general. Have I softened my stance on batteries? Am I endorsing EVs as the “right” way to mitigate our future challenges? Mostly, my answer is “no.”

I don’t hold out tremendous hope that electrified transport can smoothly replace our fossil fuel dependence. The energy density of batteries remains disappointing; most people are priced out (incentives help, but are temporary); recharging is slow and often inconvenient. What follows is some basic EV math exposing some of the hurdles.

EV Math

To illustrate some of the challenges facing electric cars, let’s consider parameters that most Americans would find to be acceptable as an equivalent trade. We’ll imagine a car that can drive a range of 300 miles (480 km): comparable to typical gasoline car ranges. Impatient Americans would like to recharge in five minutes or less. Let’s impose some hardship and say it’ll take a whole ten minutes to charge and then evaluate some of the fallout from these choices.

Charge Power and Thermal Limitations

Firstly, a person filling a gasoline tank at a rate of 0.1 gal/sec (topping off a typical tank in about two minutes) is delivering energy to the car at a rate of about 13 MW. Think about this. That’s 2,000 homes running air conditioners. Two people filling up at a gas station reaches parity with the UCSD campus’ electrical power demand. Right away you see the problem with transferring electrical energy to a car at similar rates.

But let’s get back to numbers more relevant to EVs. A 300 mile range will require approximately 80 kWh of on-board battery storage. This is based on typical EV performance demanding about 33 kWh from the wall to propel the car 100 miles (characteristic of Tesla, Leaf, Volt, C-Max, Prius; see table below, and post on EV energy efficiency), so that 300 miles demands 100 kWh from the wall outlet. At 80% charge efficiency, the battery holds onto (has a capacity of) 80 kWh. Delivering 100 kWh in 10 minutes (one sixth of an hour) demands a charge rate of 600 kW. That’s serious. We’re talking about a 2500 amp breaker at 240 VAC. Not in my house! Upscale neighborhoods beware of Tesla-induced brownouts…

But the charging problem is also bad on the thermal front. At an 80% charge efficiency, 20% is lost as heat. For reference I measure my C-Max to consistently get just 70% efficiency at 11.5 amps and 120 VAC; and 80% at 14.5 amps and 240 VAC. A 20% heat loss for our dream battery becomes 120 kW nightmare of waste heat to dissipate. Distributed over a 6 m² area (picture a cube 1 m on a side, or a flatter package fitting under the car), this is 20,000 W/m². Aggressive ventilation may achieve a convection coefficient of around 50 W/m²/°C, but this still leaves a 400 °C surface temperature above ambient. Wowzers. We have ourselves a thermal problem, folks. Partial charges at a lower state of charge may manage to be more efficient, but that’s not a full solution to the problem.

Energy Density and Mass

Modern EV batteries achieve energy densities from 0.08 kWh/kg (Nissan Leaf) to 0.12 kWh/kg (Tesla Roadster: $$). Gasoline, by contrast, packs 36.6 kWh into each gallon, with a mass of 2.77 kg, or 13.2 kWh/kg—over 100 times better than batteries. Better conversion efficiency reduces the factor of 100 down to 20–30, but still, our 80 kWh battery will be in the ballpark of 1000 kg, which is a lot to haul around.

Economics

How about the economics? Typical real costs of EV batteries today run about $500/kWh. So our 80 kWh battery costs $40,000. Actually, EV batteries only provide access to part of the capacity, lest extreme discharges ruin the battery. So our example really demands > 100 kWh of battery, pushing the cost above $50,000. And that doesn’t yet count the cost of the car. I think it becomes clear why the Tesla cars (longest range EVs) are so darned pricy. It’s more than the sleek looks and status.

Driving 100 miles in the C-Max Energi takes either 34 kWh from the outlet or 2.3 gallons of gas. At prices of $0.15/kWh and $4.00/gal (California), the propulsion cost is $5.10 vs. $9.20 to drive on electric vs. gasoline. Saving about $4 per 100 miles driven translates to $5,000 of propulsion savings over a 125,000 mile presumed lifetime (battery longevity). But we paid a price for the battery ($40,000 in the 300-mile-range case). If we want to break even, we need the battery cost to be below $5,000—meaning less than 10 kWh of on-board battery. Most EV batteries only let you use about 70–80% of the full capacity, which is almost perfectly offset by the charging efficiency. The net effect is that a 10 kWh battery (allowing use of 8 kWh, say) will demand 10 kWh from the wall to recharge, and this gets you 30 miles down the road—using our EV-constant of 33 kWh/100 mi. 30 miles (48 km) is a bit short for an electric-only car. So electric-only cars are at present not likely to break even financially.

From past experience, I expect to be attacked by vested EV owners on this point. Let me first say that there are good and valid reasons to own an EV beyond the narrow dollars and cents perspective. Bravo. But also, incentives and marketing ploys can distort the true “here and now” costs. For example, Chevy lists a replacement battery (16 kWh) at a price somewhere around $2,000—far short of the estimated $8,000 cost. An auto industry executive/scientist once told me when probed on this point: “We never discuss cost; only price.” The point is that fears of premature battery failure and high replacement cost can damage sales, and therefore place investment in the product line at risk. The industry is smart to low-ball battery prices to keep sales moving, knowing that very few replacements will be needed in the near term. If you want to experience the difference first hand, try to get a $2,000 price guarantee 10 years into the future from Chevrolet. Ford quoted me a $4,400 price to replace the 7.5 kWh C-Max Energi battery—in line with the “expected” cost of batteries.

Since pure EVs have trouble breaking even on propulsion cost, what about PHEVs, which can tolerate shorter electric ranges? As soon as a portion of the miles are driven in gasoline mode, the propulsion savings erodes, translating into a diminished break-even battery size. A vicious cycle begins, wherein each battery/range reduction translates to greater gasoline reliance and therefore further-diminished savings. Maybe the battery lifetime increases as well with lighter use. But this depends on driving profiles, number of cycles, etc. It isn’t clear that there is a pure financial win on the PHEV side, either (plus the PHEVs are more complex than EVs, driving up the non-battery portion of the cost).

My Take-Away

Despite “buying in,” I remain unconvinced of the degree to which EVs will revolutionize transportation. Don’t get me wrong: I am very satisfied with our PHEV. We went 1304 miles on our first tank of gas, lasting 58 days (697 of 700 around-town miles were on pure electric; gasoline was used for 220 miles of a round trip to Pasadena, plus the first 380 miles of road trip). I now have a car that I can charge from my off-grid PV system even if long gas lines coincide with a power outage. I love the freedom and versatility. And the car is (for me) a leap into futuristic technology that is very nice, albeit causing me some discomfort as someone who prefers simplicity (e.g., still got the flip phone) over pizzazz.

But this luxury car is just that: a luxury. It’s a novelty; a toy. Sure, it serves a purpose, and brings pleasure/independence. What works for us, though, does not mean our car-crazed civilization is ripe for an EV revolution. Limitations on charging times, range, cost, materials, and battery life may not permit a business-as-usual substitution of gasoline cars with EVs. We may not find the prosperity to pull it off. We may one day look on the car era as a carefree anomaly.

Owning an efficient, data-rich PHEV has changed my outlook on transportation and also my driving habits. Slow, complete stops give me quantitative brake-coach feedback. Modest accelerations and cruise speeds let me stretch the miles. Careful route planning and consolidation reduce the number of trips and optimize charge schedules. I do not take mobility for granted to the extent that I did. Just like in my off-grid PV system, the energy becomes more personal and precious. And that’s a good shift. We could all use more of that, in my opinion.