Mileage and range benchmarks

Hi,

A decade of Prius ownership led me to understanding how the EPA metrics work versus ad hoc measurements by individuals using often using borrowed press cars and stunts.

A name from the past, Wayne Gerdes of CleanMPG was one of the early 'stunt' drivers who deliberately drove hybrid and any ICE car to achieve exceptionally high mileage. The penultimate was Jeremy Clarkson of old Top Gear who staged a mileage race between a BMW M3 and Prius in which the BMW 'won'. A race driver in the Prius drove the track doing maximum accelerations and braking probably in "B" mode followed, by the BMW M3 on cruise control at a steady seed. Clarkson claimed 'mileage is how you drive the car.' In contrast, my interest has always been in the operational characteristics of our cars:

Source: http://home.hiwaay.net/~bzwilson/prius/

• Knee in the curve - the 2001-03 Prius achieved 52 MPG between 50-65 mph.
• Slower than 50 mph, the car lost energy switching between hybrid and engine running mode.
• Above 65 mph, the mileage fell off from higher drag that forced the engine to run all the time.
• Move knee - the 2004-2009 Prius moved the knee to 85 mph and exhibited no low speed, control issues.

Knowledge about how the car works means the driver can achieve optimum performance. In my case, 52 MPG in the 2003 Prius. Yesterday, I applied the same benchmark discipline to tow a trailer, round trip 200 miles, to Murfreesboro TN with our BMW i3-REx. Ordinarily a 70 mph car, 24x7, it could only go the distance at 50-52 mph due to trailer drag. What does the EPA have to do with benchmarking car and EV performance?

The key EPA metric are the roll-down coefficients to calculate the drag power:

drag HP = v * (A + (B*v) + (C*(v*v))) / 375

v - speed in MPH
A - drag force at any speed built-in to the vehicle
B - drag force proportional to rolling speed
C - drag force proportional to aerodynamic speed
1/375 - scaling factor to derive HP (746 W)

Note:

• drag force - (A + (B*v) + (C*(v*v))
• drag HP/kW power - v * (A + (B*v) + (C*(v*v))) * K_1
• drag energy Wh - (v*v) (A + (B*v) + (C*(v*v))) * K_2

​

The EPA applies an 'engineering' correction factor to the calculated MPG/MPGe to determine if a car maker may have a problem: https://www.epa.gov/enforcement/hyundai-and-kia-clean-air-act-settlement

(Washington, DC - November 3, 2014) – Automakers Hyundai and Kia will pay a $100 million civil penalty to resolve alleged Clean Air Act violations based on their sale of more than 1 million vehicles that collectively will emit approximately 4.75 million metric tons of greenhouse gases (GHG) in excess of what the automakers certified to the EPA. The companies will forfeit GHG emission credits in order to put the companies in the place they would have been had they accurately reported the GHG emissions from these vehicles in the first place. The companies also will take measures to prevent future violations. ...​

Companies can 'sandbag' (i.e., reduce their reported MPG/MPGe) to avoid risking a fine. But over claims can be an expensive problem ($100 m.) So we can expect EV manufacturers to report worse MPGe via the EPA. This allows ad hoc drivers and reviewers to report how much better than EPA they got. But here is how the EPA numbers do their best work:

Given a choice between EPA metrics and an opinion, I prefer engineering. The EPA has made at least two changes since 2005 including our 2003 Prius:

I do like head-to-head comparison drives where two or more cars take the same route swapping drivers. The best example: https://www.edmunds.com/fuel-economy/gas-sipper-smackdown.html

These days there are a lot of small cars pounding their chests and claiming they get the best gas mileage around. Well, we got tired of the talk and decided to settle the score. Right here, right now!

We picked three cars from our long-term fleet known for squeezing extra miles out of every gallon of gas: the spunky 2008 Smart Fortwo with a 1.0-liter, three-cylinder engine, the thrifty 2008 Ford Focus SES coupe with a 2.0-liter four-cylinder engine and a 2005 Volkswagen Jetta TDI, with its legendary 1.9-liter "TDI" diesel engine. Then we added a heavily favored gasoline-electric hybrid to the mix: a 2008 Toyota Prius. ...​

Just single car, single driver, one day benchmark tends to make my eyes glaze over. Like the old Wayne Gerdes claims, there are a lot of opportunities to accidentally screw up their benchmark.

Well so much for my rant. I prefer: EPA engineering metrics; then 'honest' head-to-head benchmarks, and; least ad hoc single car, single reviewer, and single benchmark reports.

Bob Wilson