They got an SUV and attached a sensor that estimates mileage by measuring air flow to the engine. They did 15 laps around a track, with a constant speed of 45 MPH, measuring mileage as they went. They did 5 laps with the AC off and windows down. 5 laps with AC off and windows up. 5 laps with the AC on and windows up. No laps for AC on and windows down. Here are the results.
- AC off, windows up = 11.7 MPG
- AC off, windows down = 11.3 MPG
- AC on, windows up = 11.7 MPG
They get two SUVs and emptied the gas tanks. Then filled both tanks with exactly 5 gallons of gas. They both drive around the track, same loads, this time at 55 MPH. One SUV has AC off and windows down, while the other has AC on and windows up. The AC on and windows up SUV stopped first. The AC off and windows down SUV went a 15 percent greater distance.
So, if you trust this test, AC burns 15 percent more gas. Windows down is the way to go. They end by saying they did not really trust the first test because it was just an estimate. The second test was full proof and show the best option for saving fuel and keeping cool was windows down and AC off. Where to start with the mistakes? First, you don't change your testing methods because you don't like the results.
It's really the first rule of the scientific method. You have to attack the method of estimating mileage with air flow. You can't just say, it did not do what I wanted, so it's useless. What's the point of an experiment if you can do that? Didn't they look into if air flow could measure mileage before hand? They should do experiments or research on the how sensitive air flow measurement is in estimating mileage changes.
Changing the speed to 55 MPH in the second test is just inexplicable. They had enough problems going to a different way of measuring mileage and not doing enough experiments. In the middle of all of that, why throw in another variable? They really needed to do a lot more trials and be a lot more careful. I really don't have much hope for their methods in the revisit. These are really basic mistakes.