Ex-Tesla Spokesman: EPA's Electric-Vehicle Range Figures Are Grossly Inaccurate

Ex-Tesla Spokesman: EPA's Electric-Vehicle Range Figures Are Grossly Inaccurate

By Scott Doggett May 18, 2009

Darryl Siry (left ), Tesla Motors' former chief marketing officer, blogs that he sees a troubling pattern emerging in how the most critical aspect of electric vehicles - range - is discussed by companies and the media alike.

"These are problems that could have a significant negative effect on the way the public responds to electric vehicles if manufacturers don't change the way they communicate expectations about range," wrote Siry, who is now a senior analyst for clean technology at Peppercom, a strategic communications firm.

The basic problem is that when an EV is described, it usually has a single "range" number associated with it, Siry wrote.

For example, the Tesla Roadster has a range of 244 miles. When people talk about the range of a car that is planned in the future, they also offer a single number. For instance, the media has reported that several car companies plan to come to market with EVs that have "a 100-mile range."

"Every time a single range figure is given, it should have about three asterisks next to it," he wrote.

'Your Mileage May Vary'

The biggest of the asterisks is the EV equivalent of "your mileage may vary." Using the Tesla Roadster as an example, Siry wrote, the car can indeed achieve a range of 244 miles (which is the Environmental Protection Agency's "combined number".)

In fact, one Roadster was driven as far as 275 miles before it was fully depleted, he noted. But under aggressive driving the actual range from a full charge to completely dead can be dramatically lower.

That is all well and good, but the problem is that the EPA driving cycle numbers systematically overstate what the typical driver is going to see in their daily driving.

It wouldn't be so bad if the EPA number was close to the average and depending on your driving you might see less or you might see more, wrote Siry, but it doesn't work that way. In reality, the EPA number is essentially an upper-limit number.

"The actual range you will get from a complete charge depends on a lot of factors, but I would say that as a general rule of thumb, if a company quotes an EPA range, you should apply a factor of 70 percent to that to get a realistic average range for a full charge."

'Full Charge'

The second asterisk relates to how "full charge" is defined. You'd think that should be straightforward, but it's not, he writes. In general, a battery pack should not always be charged to it's peak nor should it be drained to completely (or even nearly) empty.

"This is generally bad for the longevity of the pack. However, when the EPA test is done, the battery can be charged to its absolute maximum and the car is run until the wheels literally stop moving. This is not how a typical customer will experience a 'full charge,' " he wrote.

The typical charge settings for a car will charge a battery when it's still 10-15 percent energized. It may even limit performance or go into limp-home mode at some level above that, perhaps with 20 percent charge remaining.

These are all important factors to consider when you assess the realistic range of an EV, Siry wrote. "Depending on how the manufacturer has designed the battery management system, and depending on how the EPA test was conducted, you may have to apply another 70-80 percent factor to the range that the manufacturer states.

"Combine factors one and two above, and you are talking about average usable range of the car potentially being half of what is quoted as the EPA range. That is a very big gap in expectations that will come home to roost with consumers. If you think 'range anxiety' is a big issue, wait until the average consumer buys the car and on day 1 the average usable range is about 50-75 percent of what they were told in the marketing material (depending, of course, on how aggressive the marketing claims are)."

The Two Ranges

The third asterisk relates to the EPA's "beginning of life" range. The problem with it, as Siry wrote, is that the maximum capacity of a battery pack dwindles over time.

There are a lot of factors that affect how rapidly this reduction in capacity occurs, including number of cycles (roughly this can be expressed as miles driven/total miles per full charge), absolute temperature, variability of temperature over the life of the pack, average depth of discharge.

"This can also be a complicated discussion, but suffice it to say that the effects can be significant," Siry wrote. "It isn't an exaggeration to say that you should expect that the range of your EV could be 20 percent less after five years of use. In fact, that's being charitable."

Siry wrote that this is especially true for certain chemistries or cell types, such as the high-energy cobalt oxide cells used in laptops and cell phones. The end-of-life range of these types of packs could be significantly lower and the reduction much more dramatic. This is one reason that almost all manufacturers are moving toward chemistries that exhibit better cycle-life qualities.

So building on the example above, the realistic end-of-life range of an EV may be well below half of what was advertised.

This is a serious issue because the general public is not going to easily understand all the mental gymnastics that go into having a good understanding of what to expect from an EV, Siry wrote. This is also an issue that gets more serious as EVs go mainstream and are no longer purchased mostly by wealthy early adopters willing to forgive these quirks and inconveniences.

Solutions 1 & 2

He wrote that there are two solutions to this problem - and mass adoption of EVs by the mainstream public depends on them.

The first is that manufacturers need to communicate honestly and transparently about the realities of range. The second, he wrote, is that the EPA must establish new guidelines and standards specifically for EVs that address the issues outlined above.

Most importantly, the EPA should develop a standard that includes both the beginning-of-life range and the end-of-life range with a common definition of the expected life of the vehicle.

"If both of these things happen, we can avoid the consumer backlash I fear we are headed for with regard to range expectations," Siry wrote. "With so much progress being made on the EV front, I would hate to see the momentum slowed by false promises and disappointed consumers."

Scott Doggett, Contributor

Scott Doggett:  is an AutoObserver.com Associate Editor.