EPA to tell drivers real-world mileage...

[Ryan]

Originally posted by: Kelvrick

Originally posted by: Ryan

I'll highlight the section above. Running a car in a test facility/enclosed room in no way compares to actual driving conditions.

That was MY brightest idea that I later admitted was stupid because I was at the end of my 14 hour shift, not how THEY test cars.

Maybe I'm thinking on too grand a scale or something, but I just kept thinking about variables in weather and driver ability, habits, and just day to day differences (until they come up with a robot that can drive the cars the same every single time, which I dont' think they have yet) and would prefer to have tests under the same conditions for my own comparison.

I have yet to see mention of a good way to do these tests without said variables in this thread.

Personally - here's what I would propose:

- Build a track facility where a vehicle could travel at a constant speed of 70 mph (the speed limit on nearly EVERY major highway in the US - not this 49-55mph bullsh!t).

- Dyno the car in a wind tunnel simulating real world drag from air.

- Run air conditioning for mixed intervals.

- Calculate MPG by real world gas usage (gallons used/miles traveled) - don't measure emissions to calculate MPG figures as they currently do (one reason the Prius does so well in this test is because it's motor doens't run at all for a large part of the test).

- In city driving - mix in more real world driving conditions: Accelerate faster, brake harder.

 

[BaliBabyDoc]

Arguably, the most troubling aspect of this "reform" is that the revised MPG rating will not be used for CAFE. I wonder how long it will take Sierra Club to sue?

 

[dullard]

Originally posted by: Ryan
Personally - here's what I would propose:

- Build a track facility where a vehicle could travel at a constant speed of 70 mph (the speed limit on nearly EVERY major highway in the US - not this 49-55mph bullsh!t).

- Dyno the car in a wind tunnel simulating real world drag from air.

- Run air conditioning for mixed intervals.

- Calculate MPG by real world gas usage (gallons used/miles traveled) - don't measure emissions to calculate MPG figures as they currently do (one reason the Prius does so well in this test is because it's motor doens't run at all for a large part of the test).

- In city driving - mix in more real world driving conditions: Accelerate faster, brake harder.

Good start, but can I please add to that list with stuff I mentioned above:
[*]Have the government test the vehicles, not the manufacturers.
[*]Have the test be done on production model cars, not one-of-a-kind vehicles designed for better fuel efficiency.
[*]Use gasoline that people can buy, not a special unique blend.
[*]Use ~38% highway driving to reflect real world usage.
[*]Add some hills.

 

[Ryan]

Originally posted by: dullard

Originally posted by: Ryan
Personally - here's what I would propose:

- Build a track facility where a vehicle could travel at a constant speed of 70 mph (the speed limit on nearly EVERY major highway in the US - not this 49-55mph bullsh!t).

- Dyno the car in a wind tunnel simulating real world drag from air.

- Run air conditioning for mixed intervals.

- Calculate MPG by real world gas usage (gallons used/miles traveled) - don't measure emissions to calculate MPG figures as they currently do (one reason the Prius does so well in this test is because it's motor doens't run at all for a large part of the test).

- In city driving - mix in more real world driving conditions: Accelerate faster, brake harder.

Good start, but can I please add to that list with stuff I mentioned above:
[*]Have the government test the vehicles, not the manufacturers.
[*]Have the test be done on production model cars, not one-of-a-kind vehicles designed for better fuel efficiency.
[*]Use gasoline that people can buy, not a special unique blend.
[*]Use ~38% highway driving to reflect real world usage.
[*]Add some hills.

I agree with all of those points - there needs to be an independent body doing the testing. And hills are a good thing (I forgot about them - we don't have those in Floriduh 😛 )

 

[Kelvrick]

Originally posted by: dullard

Originally posted by: Ryan
Personally - here's what I would propose:

- Build a track facility where a vehicle could travel at a constant speed of 70 mph (the speed limit on nearly EVERY major highway in the US - not this 49-55mph bullsh!t).

- Dyno the car in a wind tunnel simulating real world drag from air.

- Run air conditioning for mixed intervals.

- Calculate MPG by real world gas usage (gallons used/miles traveled) - don't measure emissions to calculate MPG figures as they currently do (one reason the Prius does so well in this test is because it's motor doens't run at all for a large part of the test).

- In city driving - mix in more real world driving conditions: Accelerate faster, brake harder.

Good start, but can I please add to that list with stuff I mentioned above:
[*]Have the government test the vehicles, not the manufacturers.
[*]Have the test be done on production model cars, not one-of-a-kind vehicles designed for better fuel efficiency.
[*]Use gasoline that people can buy, not a special unique blend.
[*]Use ~38% highway driving to reflect real world usage.
[*]Add some hills.

I agree with most everything, but how do you propose they move this test vehicle at 70mpg around the test track? Plus, I think the limit is 65 on most, at least in California, although everyone travels at around 70.

Iono, but I just think people are in much too many different conditions to just add hills and AC. I mean, in sacramento (where it doesn't get HOT HOT (100 in summer), or COLD COLD (lowest is high 20's)), but I only run ac in the summer basically, or to help defrost in the winter, but I know in summer I'm going to get worse mileage because so. Hills, how steep of hills. What grades up and down, what percentage? Constant uphill, downhill?

I don't think highway and city should be mixed, there should be a pure highway and a pure city number and people can mix it themselves.

The last thing is funding. I'm just guessing, but that might be the reason they're having the manufacturer test the cars. Although I guess a good way is to purchase from randomly selected dealers through the country (say 5 of each car, then average the numbers) and then have the manufacturer reimburse.

 

[Ryan]

Originally posted by: Kelvrick

Originally posted by: dullard

Originally posted by: Ryan
Personally - here's what I would propose:

- Build a track facility where a vehicle could travel at a constant speed of 70 mph (the speed limit on nearly EVERY major highway in the US - not this 49-55mph bullsh!t).

- Dyno the car in a wind tunnel simulating real world drag from air.

- Run air conditioning for mixed intervals.

- Calculate MPG by real world gas usage (gallons used/miles traveled) - don't measure emissions to calculate MPG figures as they currently do (one reason the Prius does so well in this test is because it's motor doens't run at all for a large part of the test).

- In city driving - mix in more real world driving conditions: Accelerate faster, brake harder.

Good start, but can I please add to that list with stuff I mentioned above:
[*]Have the government test the vehicles, not the manufacturers.
[*]Have the test be done on production model cars, not one-of-a-kind vehicles designed for better fuel efficiency.
[*]Use gasoline that people can buy, not a special unique blend.
[*]Use ~38% highway driving to reflect real world usage.
[*]Add some hills.

I agree with most everything, but how do you propose they move this test vehicle at 70mpg around the test track? Plus, I think the limit is 65 on most, at least in California, although everyone travels at around 70.

Iono, but I just think people are in much too many different conditions to just add hills and AC. I mean, in sacramento (where it doesn't get HOT HOT (100 in summer), or COLD COLD (lowest is high 20's)), but I only run ac in the summer basically, or to help defrost in the winter, but I know in summer I'm going to get worse mileage because so. Hills, how steep of hills. What grades up and down, what percentage? Constant uphill, downhill?

I don't think highway and city should be mixed, there should be a pure highway and a pure city number and people can mix it themselves.

The last thing is funding. I'm just guessing, but that might be the reason they're having the manufacturer test the cars. Although I guess a good way is to purchase from randomly selected dealers through the country (say 5 of each car, then average the numbers) and then have the manufacturer reimburse.

The car manufacturer should provide the cars for the EPA testing. As for the testing facility - the dyno would work just fine for testing the 70mph highway MPG figures I guess - as long as it's in a wind tunnel to simulate real world drag, and maybe some mechanism to simulate resistance can be integrated into the dyno to simulate hills).

Sure - the details of which ratio of hills and air conditioning would have to worked out, but that's still better than an unrealistic vacuum environment that produces unrealistic figures.

 

[Kelvrick]

Originally posted by: Ryan
The car manufacturer should provide the cars for the EPA testing. As for the testing facility - the dyno would work just fine for testing the 70mph highway MPG figures I guess - as long as it's in a wind tunnel to simulate real world drag, and maybe some mechanism to simulate resistance can be integrated into the dyno to simulate hills).

Sure - the details of which ratio of hills and air conditioning would have to worked out, but that's still better than an unrealistic vacuum environment that produces unrealistic figures.

So it seems we're both agreeing that current EPA testing sucks, but just aren't agreeing on how far they should go towards "real world" results. I think I'd rather have the manufacturers reimburse for cars purchased for testing instead of provided cars though.

EDIT: You have to agree though, no matter how unrealistic these tests may seem, aside from big suv's in congested traffic and hills and the hybrids, you can get the results in the real world.

 

[jagec]

I think they should also move from a "mpg" to a "L/100km" (or at least "gal/100 mile") rating, since mpg figures are VERY misleading in practice. One "mile per gallon" is worth a lot less at higher MPG ratings, even though people don't think of them that way. For example, people who get a rating of 15mpg aren't too worried about only getting 14mpg, whereas someone rated at 40mpg would complain all day long about 35mpg, even though the difference in fuel consumption is lower than the first guy.

 

[Linux23]

i always thought the Govt. did the EPA tests.:Q