Nice!


So was the LOM reset each tank full?
Or is it showing the cumulative for the whole trip?

 

Reset for each tank.

 

I have never understood why people assume the pump's fuel meter is perfectly accurate but the truck's is inherently flawed. the 5% difference between the two is consistently attributed to a problem with the vehicle.

 

Good point. And it might not even be accidental error. I bet if the computer calculations were consistently reporting mpg less than actual, they would fix it.

 

Because the pump's meter is inspected frequently by the government to ensure it is accurate.

 

true, but most states have a tolerance of 2.5%, and at least here in TX they don't issue a citation or an "out of order" tag unless the pump is found to be twice the tolerance out of spec. so they could be feasibly 4.5% off and still be considered in working order per those inspections.

 

My searching has found a couple of things on this issue. First, the national allowable error is .3%, as explained here: https://www.quora.com/How-accurate-a...n-gas-stations. Further, that unless messed with, as in removing the calibration pin, which is easily detected, the pumps will be that accurate.

Second, that while .3% is the national standard, each state sets the acceptable range of error, and most appear to have adopted .5% as the limit. Car & Driver says Michigan's acceptable error is .5%: The Pump Police - Feature - Car and Driver. And this says the acceptable range in Texas is also .5%: Think the gas pump is wrong? You may be right - Houston Chronicle.

So, I don't know where you got your information but I don't think it is correct.

 

I got my information from working with people in the industry, but here's some verification from the state:

https://www.texasagriculture.gov/Reg...erReasons.aspx

The Tolerance is 6 tablespoons out of 5 gallons, or 0.5%. It looks like my info was wrong - perhaps they based it on 1 gallon? I dunno. 2.5% is what they threw around as if it was just the accepted standard.

Still, there are so many variables that I think the point remains true. ambient temperature, fuel mix, pumping technique, etc... all can cause the numbers to skew one way or another. Not to mention that the miles calculation itself is not completely accurate. as tires wear and road conditions change, so must the mileage calculation... but of course, it cannot.

I guess I'm just trying to say that a 5% difference is perfectly acceptable. calling a gauge a Lie-O-Meter and "proving" it by using so many other flawed instruments is quite silly IMO. You cannot claim that you manually calculated something when all you did is use different automated calculators. Dividing one number by another yourself does not somehow introduce a new amount of legitimacy to the resulting figure.

 

If the odometer is being used to tabulate mileage, tire wear and road conditions do not play a factor in the calculations, because that is the exact same mileage the computer is using.

And gas pumps that are inaccurate do not always dispense less gas than indicated, but can also dispense more gas than indicated. Which further throws into question why the computer calculations (as reported here) consistently indicate higher MPG vs manually calculated, and none reported less. Why is there no variation?

Here is an article from a Wisconsin study, where most of the faulty pumps dispensed extra gas:

Gas pumps not always accurate

 

All of the "so many variables" impact the LoM"s calculation the same as a manual calculation - with the exception of "pumping technique", meaning how full I get the tank each time. While I am very consistent with how I do it, I'm sure that how much I get it varies from tank to tank. However, that averages out completely when you measure and record every tank, as I have done.

As for "using so many other flawed instruments", what might they be? We just agreed that the gas pumps are accurate to .5%. The only other "instrument" used in the calculation is the truck's odometer - which is used in both the manual calculation as well as the LoM's.

But, you are right on one point - I did not manually calculate my mileage. I put the gallons used and the miles driven into a spreadsheet. And, while it is remotely possible that the spreadsheet is wrong, it does get the same results as my AccuFuel app on my iPhone.

All I'm saying is that the LoM has the ability to be extremely accurate. But most, if not all, people that take the time to record how much fuel they use across several tanks find that it is optimistic by around 5%. And always optimistic. That cannot be a fluke.

 

I guess I'm one of the rare cases, then. on my 2011 F-250, I never had a variance more than 2% off of pump calculations, and it was not always optimistic. The same was true for my 2015 Tundra. My wife does not log her fuel regularly (neither do I, anymore) but her 2011 explorer was very accurate on the few long trips we took, as is her 2015 Optima. I have not, and likely will not, continue to log fuel since I have always found the onboard computer to be close enough for my needs.

At the end of the day, why is a 5% difference even an issue? Who cares how accurate the number is, as long as it's consistently inaccurate? If you know you average X, and you want to save some fuel you shoot for X + Y. Leave the real measurements to people willing to take them using accurate methods. Weigh the fuel tank before and after, etc. Unless you're doing what it takes to be truly accurate, why is it worth arguing about how accurate something is?

 

Good points I guess, but a 1 MPG difference on a 36 gallon tank comes to 36 miles difference on a 36 gallon tank. If the console is saying you are going to get 720 miles on a tank, and you're only getting 684 miles (examples I realize), then that could be an issue.

Now I never run the tank to the point that I'm worried about whether I'll "make it" or not. When I see ~~ 100-150 miles left to go, I am going to be looking for someplace to refuel.

I will point out that there are still only two variables involved. Miles and gallons. If the console & hand calc are the same (and they are), then the only difference is the number of gallons.

If it's consistently off by the same amount, then that should be a correctable number.

 

Here's the spreadsheet I used on my recent ~2400 mile trip towing the boat, and there are several things to note. First, the LoM was off by exactly 5.00% for the trip. However, its error varied from 2.75% to 6.55% on a per-tank basis. And, all of the errors are positive.

From this I think that my truck's LoM would benefit from a 5% adjustment. However, that means that there would have been tanks on this trip where it was still off by a bit over 2%. But, on the average it would have been dead-on.

Having said that, 5% might not be correct for your truck. And, if GlueGuy finds out how to adjust the LoM, you'll have to figure out what adjustment your truck needs by tracking several consecutive tanks like I did.

 

I dont worry about the LOM. I just figure about a mile to gallon less than what it says, and thats close enough for me. like whats been said above, too many variables.......In my case, the LOM says I usually get about 15, I figure 14. and this is on a 2013 with the 6.2.

 

I (finally) pulled the notebook out of the truck and entered the data into my running spreadsheet. I now have a little less than 15,000 miles, and I've kept track since the truck was new. I do have some miles pulling a small travel trailer, and a couple of other utility miles with smaller trailers. That data is buried within, and I don't really remember when/where that was. However, the big dips in the MPG are probably a big clue.

First up is the actual versus indicated MPG:



The next thing is the percent error. Plot lines marked with a "dot" are the actual percent error. The sloping blue line indicates the running average. I suspect that there wouldn't be the upslope if I threw out the starting value, which was artificially low because we were starting with only partial data.