Aero Drag and EPA mpg

[aerohead]

Darin brought this up at the 'tire width vs drag ' thread and I thought I'd do a dedicated thread for broader access.
The reference I have which deals with this is:
"Tractive-Energy-Based Formulae for the Impact of Aerodynamics on Fuel Economy Over the EPA Driving Schedules"
by Gino Sovran,General Motors Research Laboratories
SAE Paper # 83034,1983
--------------------------------------------------------------------------
The mathematics is very contextual,relying on a 'Representative' 1983 car of Cd 0.475,Frontal area 1.98 sq meters (21.3 sq-ft),Mass 1,279 kg (2,819 lb),Turbo-hydramatic transmission,radial tires;with 26 mpg Urban,41 mpg Highway,and 31 mpg Composite.
From Sovran,"There is no intent to develop a means for predicting actual on-road fuel economy level per se.Furthermore,the relationship between on-road and EPA dynamometer fuel economy is not addressed."
--------------------------------------------------------------------------
Sovran develops an 'Influence Coefficient' which takes in Drag,BSFC,Driveline efficiency,Gear-matching,Cooling fan,Alternator,Power steering,Idling,Braking,Closed-throttle,Vehicle-to-vehicle repeatability,3-speed automatic transmission.
-------------------------------------------------------------------------
Figure 3,Page 7 provides a table of mpg based on successive 10% drag reductions,for Urban,Highway,and Composite.
Here are some values for the Highway cycle with extrapolation out to a 65% drag reduction:
*0% = 41 mpg
*10% = 43.4 mpg
*20% = 46.5 mpg
*30% = 49.51 mpg
*40% = 52.11 mpg
*50% = 57.5 mpg
*60% = 63.25 mpg
*65% = 67.5 mpg
-------------------------------------------------------------------------
The Federal Test Procedure for EPA "highway" mpg consists of a 10.2-mile 'trip' on a dynamometer,over 765-seconds,with:
* 1.62-mile 'Local' portion run at an ave. 40.7 mph
* 2.10-mile "Collector' portion run at an ave. 43.8 mph
* 3.96-mile 'Principle Arterial' portion run at an ave. 56.1 mph
*2.52-mile 'Minor Arterial' portion run at ave. 48.2 mph.
The entire 765-second test averages 48.2 mph for the composite test.
*Maximum velocity is 60 mph during the Principal Arterial portion.
*there is one complete stop during the test.
*Reference: 1972 Federal Test Procedure (FTP)
Passenger Car Fuel Economy During Non-Urban Driving
Thomas C.Austin,Karl H.Hellman,& C.Don Paulsell
Environmental Protection Agency
SAE Paper 740592,August,1974
-------------------------------------------------------------------------
From Sovran's own words we understand that these relationships are limited to the dynamometer and do not present a useful tool to us as modders.

[2000neon]

That's interesting, I have always wondered how it is actually done. How is fuel consumption monitored? Is it just a theoretical, calculated number in the influence coefficient, or is it a measured value?

I suppose it wouldn't be practical, or easily achievable, but it would be cool to see an ecomodded vehicle put through the same test. You could have a real, variable controlled test of MPG to get an exact gain over the factory vehicle. Although aero mods wouldn't show a gain on the dyno, unless you could factor in the decrease in Cd into the influence coefficient.

[jime57]

This is very interesting. The first 10% drag reduction provided approx 5% better fuel economy, but the relationship looks to be exponential. The 10% reduction from 50-60% provides better than a 10% reduction in fuel economy.

I guess that one might infer that drag reduction is most rewarding for cars that already have low drag or Cd?

[SlideWRX]

Originally Posted by 2000neon

That's interesting, I have always wondered how it is actually done. How is fuel consumption monitored? Is it just a theoretical, calculated number in the influence coefficient, or is it a measured value?

All the exhaust from a test is captured in bags, so they can measure exactly how much CO2 the vehicle produced in a run, and therefore how much gas.

How Vehicles Are Tested

For vehicles using carbon-based fuels (e.g., gasoline, diesel, natural gas, etc.), a hose is connected to the tailpipe to collect the engine exhaust during the tests.
The carbon in the exhaust is measured to calculate the amount of fuel burned during the test. This is more accurate than using a fuel gauge.

[jakobnev]

Originally Posted by jime57

The 10% reduction from 50-60% provides better than a 10% reduction in fuel economy.

10% from the original drag is 20% of the new reference that only has 50% left. (% vs %-points and all that : )

[Sven7]

Didn't they significantly change test procedures since 1972?

[euromodder]

Originally Posted by 2000neon

I suppose it wouldn't be practical, or easily achievable, but it would be cool to see an ecomodded vehicle put through the same test.

It wouldn't change much.

Aerodynamics are not taken into account at all.
Nor are faster warm up times.
RR would be a factor on the dyno rolls.

[jakobnev]

Aerodynamics are not taken into account at all.

Not even a hint of suspicion that the aerodynamic properties of the car are used to determine the resistance of the dyno rollers? (among other things like mass)

[Otto]

If aerodynamics are not part of the calculation, then the entire EPA mileage estimate is bogus. This would explain why cars seldom get the same highway mileage in the real world.

Coincidentally, yesterday I called Wikispeed to ask how they got to their 100+ mpg claims for their car, whether it was from actual highway testing on real roads in real traffic and real weather, or from an EPA wag (wild-assed guess). No answer so far.

Some lady successfully sued Honda (I think) because her car did not get nearly the claimed EPA mileage.

If people knew the real world mileage of new cars and what little savings they'd have vs. maintaining/modding their paid-for old car, they would not buy such new cars. Fr'ex, paying ~$10K extra for the latest fuel efficient car might take hundreds of thousands of miles driving to save enough on fuel just to amortize the extra cost, including interest, maintenance, depreciation, etc..

[RiderofBikes]

Originally Posted by Otto

If people knew the real world mileage of new cars and what little savings they'd have vs. maintaining/modding their paid-for old car, they would not buy such new cars. Fr'ex, paying ~$10K extra for the latest fuel efficient car might take hundreds of thousands of miles driving to save enough on fuel just to amortize the extra cost, including interest, maintenance, depreciation, etc..

"IF"... I'd say your right, and after 1 yr 5 months later from buying my Fiesta straight from the lot. I'm still kicking my butt. Shoulda, bought something cheaper IF I took a little more time to look and research(pre ecomodder). Coulda fixed my car within a paycheck at the time(blown headgasket, found engine for $300 on CL), BUT already spending $3,500 in repairs in the 1st yr of ownership(used car) and have to throw more money at it!!? Woulda, bought something used again if it weren't for that damn StressTank i had before... So, basically that was in total $6k spent on a used car in one year, and it only got like 27mpg at best, so fuel wouldv'e equaled out overall 30,000mi/yr to the cost of a new car.. Insurance was more too!!

35,800mi later, im satisfied with my purchase. I have'nt needed to do 1 thing to it other than change oil and put gas in it. Plus i don't plan on getting rid of it for the next, 10yrs AT LEAST! so after 5yr more of payments, intrest, insu. I will be THAT much more satisfied about the choice i made.

Obviously, not everyone is as bad a situation like i was... BUT im keen to buying a cheapo civic still, since me and my fiance need a 2nd car. Modding it will will be Hellava trip, anything goes!!!!