Mileage for the real world

[cjrciadt]

http://www.freep.com/apps/pbcs.dll/a...601100425/1014
Drivers would get lower -- but more accurate -- miles-per-gallon estimates on cars they buy starting with the 2008 model year under new testing standards that are expected to be announced Thursday in Detroit at the North American International Auto Show.Automakers and consumer advocates have been calling on the U.S. Environmental Protection Agency to revamp its tests, developed in the 1970s, saying faster highways and congested roads make the EPA's numbers, derived from laboratory testing, misleading. Consumer Reports said last year that among 303 vehicles it tested, 90% fell short of the fuel economy on their window sticker, some by as much as 50%.

While the changes wouldn't immediately affect the way automakers meet federal fuel economy standards, they could push more consumers toward higher-mileage vehicles in the face of roller-coaster gasoline prices.

The EPA "needs to close the gap between what consumers experience in the real world and what they see at the dealership," said Jason Mark, director of the Clean Vehicles Program with the Union of Concerned Scientists.

According to three people familiar with the proposal, who didn't want to be named for fear of damaging relations with the EPA, the agency will change its tests in two stages.

The first stage, which would go into effect with 2008 model-year vehicles, would essentially lower mileage results by a set percentage to reflect the drain of air-conditioning and other accessories.

The adjustment might reduce the results of city mileage by up to 13%, sources said, but it wasn't immediately clear.

For cars and trucks built in the 2011 model year and beyond, the agency will propose a larger change to testing that would add three tests currently used to measure emissions to the mileage standard tests, coming up with new numbers that will likely be even lower than the previous figures.

The new laboratory test would involve measuring the vehicle's emissions in cold weather, while running the air conditioner and under hard acceleration.

The EPA is scheduled to unveil its proposal on Thursday in Detroit. The agency declined to comment Monday, but EPA Administrator Stephen Johnson told reporters in November that the test needed updating.

After proposing the changes, the EPA will have 90 days for public comment before taking final action.

Under the current regime, the EPA uses two tests, one for city driving and another for highway driving. Speeds are limited to 52 m.p.h. in the city test and 60 m.p.h. in the highway version. All take place at room temperatures, and the vehicle's air-conditioning is turned off -- conditions that critics have said fall far outside modern conditions on 70-m.p.h. freeways and gridlocked city streets.

In its testing, Consumer Reports found the city mileage was the more inaccurate of the two numbers, often wrong by 35% to 50%. Hybrids had the largest disparities, averaging 19 m.p.g. less than their EPA averages.

None of the three hybrids tested by the Free Press in 2004 approached the EPA's predicted fuel economy. All vehicles use more fuel in real-world conditions than in the EPA's controlled laboratory test. The hybrids fell short by a surprisingly wide margin -- nearly 20% in the case of the best-selling Toyota Prius -- in the 300-mile comparison test.

Further testing uncovered the fact that the hybrids' fuel economy also suffers when the vehicles' climate-control systems are set for maximum air-conditioning or to defrost the windshield.

After complaints in the 1980s that the tests were increasingly inaccurate, the EPA corrected the results by reducing the estimates by 10% for city testing and by 22% for highway estimates. The agency tests only about 10% of new models, relying on automakers to use the guidelines to test their own vehicles.

Some automakers, including Toyota and Ford, have attempted to warn consumers buying their hybrid vehicles that their mileage may vary from what's printed on their window sticker.

A spokesman for the Alliance of Automobile Manufacturers, who had not seen the proposal, said automakers "are generally supportive of EPA efforts to make fuel economy testing as accurate as possible to provide customers with the best information."

Automakers use the unadjusted test figures when calculating whether their models meet federal fuel economy standards.

Several environmental groups have called on regulators to make automakers meet real-world standards.

Contact JUSTIN HYDE at 202-906-8204 or jhyde@freepress.com. Business writer Mark Phelan contributed to this report.

[Minnow101]

What's so difficult about actually putting every new model on the road and do real world highway/city and combo. driving. If we expect real world accurate mileage resultls then one needs to have real world driving. Putting the drive wheels on a dynometer in a garage is not and has never been real world testing.

[wtgkb8]

Like I have said in the past on this site, I never have ridden in a vehilce that got outside of EPA estimated mileage. If anything, I get higher than EPA estimate on hwy. Only time I get EPA city mileage is during the coldest part of the winter, city driving(constant stop/go), allowing for 15 min warm ups. I just don't see a problem with my experiences.

[Proetus]

Originally Posted by Minnow101

What's so difficult about actually putting every new model on the road and do real world highway/city and combo. driving. If we expect real world accurate mileage resultls then one needs to have real world driving. Putting the drive wheels on a dynometer in a garage is not and has never been real world testing.

There are too many factors that they would not be able to control: temperature, humidity, barometric pressure, wind, road sufaces (concrete produces less rolling resistance), traffic, etc.

The only way to produce consistent results is in a controlled environment. To have year-round access to an environment where all of those factors can be controlled the tests need to be conducted in a lab.

[SuperSS27]

Originally Posted by Proetus

There are too many factors that they would not be able to control: temperature, humidity, barometric pressure, wind, road sufaces (concrete produces less rolling resistance), traffic, etc.

The only way to produce consistent results is in a controlled environment. To have year-round access to an environment where all of those factors can be controlled the tests need to be conducted in a lab.

All of the above mention have VERY little effect on gas mileage. The biggest thing that has an effect is elevation. If you are in traffic, how can you count the engine running when you aren't moving mileage?

[SeanCasey]

Here in Southern California where most places are fairly flat, I tend to exceed the highway ratings by 2-4 mpg and tie city milage (mainly due to horrendously long red-lights).

So I guess with the new ratings I can brag about exceeding the ratings by even more?

[01cavalier]

I have never gotten significantly less mileage than the EPA ratings in the city, and the only times I get less on the highway is when I drive in very hilly areas. Typically when I go to Seattle, I get 32-33 MPG on the highway, my car is rated for 32 and I drive 75-80 typically. They seem accurate to me.

[Proetus]

Originally Posted by SuperSS27

All of the above mention have VERY little effect on gas mileage. The biggest thing that has an effect is elevation. If you are in traffic, how can you count the engine running when you aren't moving mileage?

http://www.sciam.com/askexpert_quest...F983414B7F0000
http://www.everytime.cummins.com/eve...Whitepaper.pdf

Harold Schock, professor of mechanical engineering and the director of the Automotive Research Experiment Station at Michigan State University, explains.

Additionally, the amount of drag between tires and the road is about 20 percent greater at 0 degrees F than it is at 80 degrees F.

A vehicle’s aerodynamic drag is proportional to air density. On a 70-degree-F day, the density of the air is 16 percent lower than on a day with temperatures around 0 degrees F. Although this makes little difference in urban driving, it could account for a highway mileage per gallon reduction of 7 percent on the colder day.

For every 10 degree F drop in temperature, aerodynamic drag increases by 2%.

The problem with unpredictable traffic flows is that you can not precisely replicate the test parameters:

Elevation: 400 feet above sea level 
Humidity: 67% 
Barometer: 30.03 in/Hg 
Load: 220 lbs. (Driver & cargo) 
AC: ON (using defroster in cooler temps) 
Climate Control: 72 ºF  
Winds: None  
Fuel: Per table below (114,500 BTU summer & 112,500 winter) 
RRR: 0.001144 (Road Rolling Resistance) 
TRR: 0.0016 (Tire Rolling Resistance) 
Cd: .28 (Coefficient of Drag) 
FA: 1.93 m^2 (Frontal Area meters squared) 

MPH---AMBIENT-TEMP-----MPG (US)
50-----------85----------58.91
50-----------75----------57.42 
50-----------65----------54.98 
50-----------55----------52.73 
50-----------45----------49.76-----using winter fuel 
50-----------35----------47.87-----using winter fuel 
50-----------25----------46.10-----using winter fuel 
50-----------15----------43.66-----using winter fuel 
50-----------05----------42.14-----using winter fuel

[DuSpinnst]

EPA Tests are horrid. Either you beat the MPG by 2 to 3, or you are down by 2 or 3. No one I know gets EPA.

Besides, in addition to the lab tests, there should be real world testing to either corroborate the lab tests or prove the width of the margin of error on the lab numbers.

[mjd1001]

Originally Posted by 01cavalier

I have never gotten significantly less mileage than the EPA ratings in the city, and the only times I get less on the highway is when I drive in very hilly areas. Typically when I go to Seattle, I get 32-33 MPG on the highway, my car is rated for 32 and I drive 75-80 typically. They seem accurate to me.

I agree. In every one of my last 5 new cars, I have gotten just about right on the EPA ratings.

It's a matter of how you drive. When I do mileage tests, when I drive 65-68 on the highway, and do the speed limit around town (which is how I drive most of the time) the ratings are amazingly accurate.

When I do slip into a mode of going past 70 on the highway, or being 'more aggressive' around town, my mileage goes well under the EPA ratings.

I'm happy with the ratings they way they are. Let me know what kind of mileage I'm getting if I actually drive in a way that a sensible driver would. For the other 80% of the people who insist on going 75 on the highway, or racing from stoplight to stoplight around town...just tell them they aren't going to get the mileage ratings and leave it at that.