EPA MPG Test Doesn’t Work for Hybrids
The Center for Auto Safety is the nation’s premier independent, member driven, non-profit consumer advocacy organization dedicated to improving vehicle safety, quality, and fuel economy on behalf of all drivers, passengers, and pedestrians.
By Mark Rechtin
Automotive News / November 24, 2003
LOS ANGELES — In publicity for its Prius hybrid-electric vehicle, Toyota Motor Corp. claims the compact sedan is EPA-certified to get 51 mpg on the highway, 60 mpg in the city and 55 mpg in a “combined” driving environment.
Unfortunately for most consumers, their Priuses will never come close to that performance level.
Press a Toyota engineer, and he'll admit that most Prius owners get around 44 mpg from their cars in combined driving. That's still an impressive number, but it's 20 percent less than what Toyota tells the world.
Is Toyota pulling a fast one?
Not at all. In fact, Toyota says it would prefer to let consumers know that their actual mileage will fall short of the official rating.
But they can't because government won't let them.
Dave Hermance, Toyota's executive engineer for environmental technology, said Toyota is not allowed to publicize any mileage claims other than EPA test results.
But the EPA tests are a distortion of the real world, he said.
All vehicles – from gas-guzzling Hummer H2s to fuel-sipping Priuses – fare better on the test than they do on the road. But because mileage ratings are much higher for fuel-efficient cars, a small percentage discrepancy can translate into a big mileage difference between the test results and reality.
The discrepancy stems from an outdated EPA testing procedure that was created nearly 40 years ago and that does not reflect accurately today's driving styles or routes.
“It was developed in the 1960s, when there were limitations on the test equipment at the time,” Hermance said in an interview at the recent EVS-20 electric vehicle convention in Long Beach, Calif.
“They couldn't even brake hard because the testing equipment couldn't handle it. It really is 50th-percentile driving. No one drives like that anymore.”
Hermance drives a Prius and said he typically gets between 53 mpg and 55 mpg combined. But he knows exactly how to “pulse drive” the car – that is, to accelerate briskly and get it up to speed, then mostly coast and let the electric motor handle the slight modifications needed to keep the vehicle at speed.
A survey of 750 first-generation Prius owners on yahoo.com showed them obtaining between 35 mpg and 55 mpg combined driving, with an average of 44. An early poll of 30 2004 Prius owners showed most got between 45 mpg and 49 mpg.
One devoted Prius owner in Minnesota, known as John1701a, says he averaged 45.4 mpg over nearly 60,000 miles in a 2001 Prius. He has since purchased the redesigned 2004 model and has raised his average to 47.1 mpg.
Why are the official EPA numbers so different from reality?
A bigger penalty
The EPA city-driving test simulates an 11-mile, stop-and-go trip with an average speed of 20 mph and a maximum speed of 56 mph. The trip has 23 stops and includes time for the vehicle to idle at a standstill. The highway test simulates a 10-mile trip and averages 48 mph. The maximum speed is 60 mph.
The EPA already adjusts the fuel-economy results from its dynomometer test to account for “road load” – the difference between controlled laboratory conditions and the actual road. For city driving, the penalty is 10 percent; for highways, it is 22 percent.
But certain loads, such as running the air conditioning, are not considered. Neither is cold weather, which disproportionately penalizes battery-powered vehicles more than internal-combustion ones.
Dan Harrison, manager of the vehicle programs group for the EPA in Washington, admits that hybrids are difficult to test because “there are more variables with a hybrid than with an internal combustion engine.”
For instance, hybrids also must account for regenerative braking and the load accessories place on the vehicle.
Just the same, Harrison said hybrids should get “within 15 percent” of the official fuel-efficiency rating.
Hybrids 'drive differently'
Larry Oswald, a veteran of GM's EV1 program and now CEO of DaimlerChrysler's Global Electric Motorcars venture, says running the air conditioning takes “a big-time chunk” out of a hybrid's fuel economy.
The air conditioning can work the battery-electric motor as hard as the actual driving, he says.
“People drive harder, accelerate quicker and brake faster than the EPA test cycle,” Oswald says. “They are not operating within the optimum range for the battery, power electricity and motor.”
Says Walter McManus, a J.D. Power and Associates analyst specializing in alternative powertrains: “We're all trained how to drive a gas vehicle, but a gas-hybrid drives differently, and people are not used to it.”
In J.D. Power's 2003 APEAL Study, consumers were only marginally more satisfied with the fuel economy of their Prius or Honda Civic Hybrid than they were with the base-engine versions of the comparably-sized car, McManus says. “APEAL” stands for automotive performance, execution and layout.
The Prius' 1.5-liter gas engine is shared with the Toyota Echo. The Civic Hybrid's engine is only marginally smaller than the base Civic engine.
“The people who think they are going to get 55 miles per gallon are going to get 40,” McManus says. “They realize they could get that mileage with other vehicles, and they are going to be disappointed. The main problem has been that the consumer's expectations are not met, an unfulfillment of expectations.”
A new test for hybrids?
As a result, the EPA is studying whether to adjust its test procedures for hybrids, perhaps making them run a different cycle that includes air conditioning or altering the amount of city driving factored into the combined test.
But the EPA's Harrison doubts such a change in the test matrix will be made specifically for hybrids.
He says that other high-mileage internal combustion vehicles show different mileage patterns as well in the real world because they are tuned for a specific range of operating conditions.
Once the vehicle goes outside those conditions, Harrison says, the mileage “drops like a stone.”