GM CEO Fritz Henderson with Chevy Volt
GM's announcement that the Chevy volt could achieve the equivalent of 230 miles per gallon was both praised and criticized.

Time to Dump MPG Ratings

A proposal for a common sense method to measure electric-drive vehicle efficiency.

By Bill Moore

General Motor's announcement this week that a proposed EPA drive cycle would enable the Chevrolet Volt to achieve a headline-grabbing 230 MPG accomplished its objective. Suddenly everyone was talking about the Volt again, instead of Nissan LEAF, whose own marketing mavins responded, via Twitter, that on the same fictitious drive cycle their all-electric car would achieve 367 mpg. How an all-electric car measures its performance in miles-per-gallon is not only an oxymoron, it also highlights the complete absurdity of measuring any plug-in electric vehicle by last century's fuel economy standards.

This is a problem both GM, other car makers and government regulators have wrestled with for months and have yet to arrive at a sensible solution to which the average Little League Dad or Hockey Mom can relate. MPG we understand, but it's meaningless in the context of vehicles that may not even use gasoline or any liquid fuel.

In the case of the Nissan LEAF, the problem is fairly easily resolved: the EPA would rate the car in watt hours of electricity consumed per mile or watt hours/mi (Wh/mi). Now at the moment, that means little to gasoline car drivers, but to experienced electric vehicle drivers, it's the equivalent of miles per gallon. It is a key measure of the overall energy efficiency of car in terms of the rate at which energy is extracted from the battery to propel the car down the road. Like MPG, it is based somewhat on the average energy consumption of the vehicle, influenced by driving conditions and driving styles.

The problem becomes far more complicated with the Volt, which uses two energy sources to propel the car: gasoline and electric power. Measuring the former in MPG's is straight-forward enough, but not when you try to do so in a mixed-energy mode with some power coming from the gasoline and the rest from electric power. It really is an apples vs. oranges problem.

However, there is a very simple way to address the problem. We stop measuring energy consumption by fluids consumed and instead shift to measuring the amount of carbon dioxide produced. Now we are comparing apples to apples. We know the amount of CO2 a gallon of gasoline/petrol produces: 253 grams per kilowatt hour (g/kWh). We also know the comparable emissions for electricity: coal produces 346 g/kWh, while natural gas, the lowest carbon dioxide producer among the common fossil fuels, is 206 g/kWh.

Contrary to nuclear power industry propaganda, their electricity is not carbon neutral. While they don't produce CO2 at the plant level, the upstream fuel refining process does produce carbon dioxide. A study by the German Oko-Institute found that the upstream fuel cycle generates the equivalent of 34g/kWh. Other studies place the number at 60 g/kWh, but for the purposes of this paper, we assume 40g/kWh, certainly far below that of even the cleanest fossil fuel, but those fuels also don't have the same disposal issues as those associated with spent nuclear fuel rods.

A number of years ago, a study found that giant hydroelectric dams may, in fact, produce some CO2 -- more likely methane -- from decomposing organic matter that builds up in the sediment at the base of the dam, but we do no include that assumption in this presentation. Other renewables generate little or no CO2 in the processing of producing energy, though the upstream manufacturing process will result in some CO2 production, as would the construction of any energy production system, which we also ignore in this paper. The uniquely energy-intensive character of nuclear fuel production process, however, does demand it be included.

Now we have a common ground on which to compare conventional gasoline/diesel car, hybrids, plug-ins, extended-range electric vehicles and pure battery electrics. Conveniently, ACT ON CO2, a program run by the British Government, provides its citizens with a convenient tool to determine the carbon dioxide emissions of cars available in the UK, some of which have North America analogs including the Toyota Prius and Chevrolet Cruze, which GM used as the engineering development mule for the Volt.

One of the unintended benefits of the 230 MPG public relations gambit is that it suggests to us a couple important performance figures on both the Volt and the Nissan LEAF. GM announced that over a 100 mile drive, the Volt would use 25 kWh of energy. Since the Volt will drive 40 miles using 8kWh of energy before kicking in the gasoline generator, the total electric power that would be required to achieve this distance, involving two full and one half recharge (8+8+4 kWh), which totals 20 kWh, not the 25 quoted by GM.

So, assuming GM calculated that 40 miles of the 100 trip would be under electric power and the remaining 60 would be in range-extended mode, this could account for some 17kWh of energy from gasoline (25kWh minus 8kWh) that is consumed to drive the 100 miles. There are 34.7 kWh of energy in a US gallon of gasoline. This would result in the consumption of .5 gallons of gasoline, producing a total of 4730 grams of CO2 for the sixty miles, or 79 g/mi (49 g/km). This suggests pretty remarkable fuel economy in extended range mode of 120 mpg, which isn't entirely out of the question, but seems improbable, at least from our current vantage point.

Combining the 4730 grams of CO2 created from burning the hypothetical half gallon gasoline plus the estimated 1776 grams generated by the "average" American power grid mix (222 g/kWh) means the Volt operated on a average of 65 g/mi over the 100 mile trip. The relevance of this exercise presently will become obvious.

Assuming for the moment, however, that this is how GM arrived at the 100 miles on 25 kWh of energy, what we can derive from this is they seem to believe that even in this combined mode, the Volt will get 250 watt hours per mile (Wh/mi). Doing a bit of algebra on Nissan's 367 mpg number compared to GM's hypothetical numbers results in an even more impressive 160 Wh/mi, meaning the LEAF is one-third more efficient than the Volt. This is certainly not an unreasonable number, though I wouldn't expect anyone but the most conservative drivers could attain it. EV's like the LEAF and Volt are just going to be too much fun to drive to worry about a few hundred Watt hours here and there.

Shifting to Grams Carbon Dioxide Per Mile

Given that gasoline and electricity don't mix when it comes to energy efficiency ratings, we propose adopting TWO KEY NUMBERS: Watt hours per mile(km) for energy efficiency and grams of CO2 per mile(km) for emissions. Both enable us to compare values equitably.

Case in point, the 1.8LT Chevrolet Cruze is rated in the UK at between 159-184 CO2 g/km, depending on engine options. Converting this to miles results in 255-296 g/mi. Compare this to the Volt, which based on the above assumptions, would produce a combined 65 grams of CO2 per mile, almost 200 less than the gasoline Cruze.

The Nissan LEAF is comparable to the Versa here in North America, but there is no similar model in the UK. The closest is the Note, which is rated from 119 g/km to 159 g/km. Converted to miles, this translates into 191-255 g/mile. Again, extrapolating what we believe may be overly optimistic numbers, the 160 Watt hr/mi of the LEAF could generate as little as 35.5 g/mi (22 g/km), 82% less than a Note.

Of course, much of this depends on the source(s) of the electric power that charges the car's batteries, and that varies dramatically by locale, state, region and country. The less the local grid relies on fossil fuels, the less carbon dioxide emitted. A LEAF recharged by wind, solar, hydro and geothermal will, in effect, create no carbon emissions; one powered by a coal-dominated system will, but even here the numbers are instructive.

Take two States here in America as examples. Nebraska relies on Wyoming coal for 69% of its electric power. Nuclear power is the next highest at 26%, followed by hydropower at 3% and natural gas at 1%.

By contrast, California's power mix is one of the cleanest in the country. Natural gas supplies 51% of its generation capacity. Hydro handles 18% and nuclear power meets 6.4%. "Other," including wind, solar and geothermal, produce 14% of its power. Coal is a mere 1%.

Taking these percentages and utilizing them to arrive at the amount of CO2 per kWh each contributes we arrive the at following:

Nebraska's combined grid CO emissions are 254 g/kWh

California's combined grid CO emissions are 115 g/kWh

This means that someone living in Omaha, Nebraska who drives a Volt will generate 63.6 g/mi (39.5 g/km). Driving a LEAF, instead, would create just 40.6 g/mi (25 g/km).

A similar driver in California who simply uses the local power power grid and not his own solar panels, which many do, creates just 28.75 g/mi (17.9 g/km) in the Volt and 18.4 g/mi (11.4 g/km) in the LEAF.

Compare this to the 2010 Toyota Prius, considered one of the most fuel efficient, environmentally-responsible production automobiles on the road. The UK government rates between 89-104g/km (167 g/mi), more than double the rate of the Volt being driven in a largely coal-dependent state.

By shifting the way we compare vehicle energy efficiency to one based on the amount of carbon dioxide generated through the fuel cycle, allows us to weigh the relative merits of models and drive systems. What will become important is not a confusing MPG rating or a hybrid thereof, but two easily understood -- and calculated -- numbers (once they are explained to people): Watt hr/mi and CO2 grams/mi; and the smaller both numbers, the better. A new Volt or LEAF might have the following Efficiency Ratings: 250 Wh/mi and 160 Wh/mi respectively with the usual "your economy may vary" caveat.

On the increasingly important carbon emissions front, the stickers would ideally be regionalized and reflect the local electric power mix. A California Volt would show 30 g/mi (18 g/km), for example, while one in Nebraska would read, instead, 64g/mi (40g/km).

All of this is, of course, purely conjecture. We have no way of knowing exactly how GM or Nissan actually arrived at their respective numbers. Nor do we know if either vehicle will be anywhere nearly as efficiency as our numbers suggest. We'd be prudent to error on the side of caution. However, that they did engage in this exercize of engineering-cum-marketing tit-for-tat underscores the need for a more logical approach to measuring efficiency in electric-drive vehicles. The auto industry has been gradually introducing the concept of rating engine power by both traditional -- and clearly obsolete -- horsepower and now kiloWatts. Batteries in hybrids are now being referred to by their kWh capacity. Inevitably, it will make sense to rate vehicle efficiency in a similar manner.

Times Article Viewed: 11321
Published: 14-Aug-2009


blog comments powered by Disqus