Concerning methods of measurement

[Flatland2D]

I have some ideas about how we should measure fuel economy. I believe that the most accurate way to measure mpg is with highway driving. There are simply too many variables in city driving to get consistent results without taking an average over many gallons of gas. Driving style can fluctuate on a daily basis, as well as temperature and humidity that could effect performance. Highway driving involves practically no starts/stops and speed is always kept constant eliminating a lot of human error. Because of this I think one could get by with using less gas in each test. It would take me almost a month to go through 3 full tanks of gas as required in the experiments posted. I don't know about you but I like my results immediately. Going on a joy ride for an hour or two is a nice way to measure mpg so long as you keep it on the highway. I suggest that maybe we should have the new standard for measurement be confined to something like 100 miles of continuous highway driving at 65 mph.

Now I'm not saying that every test should only involve 2 to 3 gallons of gas, but I think that will give a good idea about if the concept being tested really works or not. If gains are shown, the concept can be verified by extended tests. No need to waste gas on tests that aren't going to help.

As for measuring fuel consumed during the test, I believe the top-off method is accurate if done properly. You must fill up before and after the test at the exact same pump at the same gas station. If you used pump #3 at your local Shell, go right back to that pump at the end of the test for measurement. Different gas stations (and different pumps even) can have varying cut-off points to prevent overflow. I once tried to test fuel economy in my wife's Civic by topping off before leaving, then again at my destination about 60 miles later. The calculated gas mileage was about 70 mpg because the second pump switched off much earlier than the first one. Also, don't add any more gas once the pump automatically stops itself. Make sure the pump is pushed all the way into the fill hole. That might affect the cut-off point, too.

Just a few ideas on how to accurately measure mpg. Please share your thoughts and opinions on this. I am eager to participate in the experiments posted but it will take me a long time to go through 3 tanks of gas, not to mention inconsistent driving style, weather, and frequent changes from city to highway driving would introduce error in my results.

[SVOboy]

I'll make an example that pertains to me and I think will help out your point a little bit, then I'll contradict you, but here's my example:

So I'm in the middle of my tires test, but as it turns out I already bought new tires and hopefully I'll get them mounted wednesday, so if I do that then it'll be damn screwed on my test doing more tanks to retest the control and then retest over-inflation, that would be about three months for me. So, this makes a lot of sense. I have a interstate right near my house so I could go top off, drive for 45 minutes to and hour, catch a clover leaf and come back and it'd be a piece of cake. So that way if I had a free afternoon I could test both the tires and the control in one 4 hour period and only a few bucks. Also, I would not have to worry about constantly checking pressure and all that.


However:

For something like acetone, that has to work its way in and out of your tank, 3 tanks is prolly a good idea. But other things that do not have a lag time, your method I think is good.

[Matt Timion]

I have been thinking about this lately myself. I actually really like Ernie Rogers' way of testing MPG. he drives a straight stretch of road for 100-120 miles and then back. For him it's from Salt Lake City to Wendover (I think) Nevada. I think I may do the same test strip as him.

Highway driving such as this show gas mileage results in the best way. Set cruise control or have a target speed and stick to it. I may end up changing the "how to conduct an experiment" section to use this much easier method.

Now, there is also the concern of additives making their way through the gas tank. For that I say it's best to drive around for a few days after adding the additive, then fill up, and add the appropriate amount of the additive at the fill up before the test run. They say with acetone it takes 70 miles or so for it to work it's way though the system. Sounds doable to me.

IOW, I think you're right about the highway driving. too many variables are involved in city driving, and highway driving is always the same assuming you always use the same strip of highway, and the same speed.

[Flatland2D]

The point about acetone is a good one. I have read about it taking three tanks of gas to reach peak efficiency, others say it happened in just one tank. If anyone has an explanation for why it takes so long to see the effects I'd like to hear it. If all acetone is doing is making the gas easier to vaporize, I don't see why the effects are not noted immediately. Acetone ought to dissolve pretty quickly into a homogenous solution in the tank.

An inital break-in period might be required for certain modifications. Keep putting in acetone but don't start any tests until you've gone through three tanks of gas. I wonder if hydrogen works the same way. Anyway, it's not that this method is inaccurate, it's just that you might have to postpone the tests for a while.

Immediate results should be seen from wind drag modifications, tire change, air pressure and the like.