MPG On Screen vs Reality

[Guardian_Bob]

So I'm seeing a discrepancy between my measured MPG (based on the pump counting the number of gallons going in) and the number of gallons consumed on the Trip display.

 

My next experiment (in about 3 weeks, after I have the fuel thing fixed) will be to reset the trip counter and check after 5 fill ups.

 

That said I don't want to do this alone, as one data point does not make a trend. Is anyone willing to check the number of gallons the pump reports going in over 5 fill ups and the amount of gas consumed in the trip computer?

[jeff_h]

Mine is perpetually about 1.5-2 MPG less in actual miles/gallons at the pump... today was the largest difference I've seen as I filled up with 591 miles, Trip 1 (which I reset with every fill) said 52.7 however my actual was 49.03 however I packed the tank completely full and parked on slight incline (last fill was parked in very slight decline, so some air could have been trapped in part of tank)... so the car tends to be about 1.5-2 MPG optimistic on the display on the tanks as compared to the actual -- but I have also noted in 1 or 2 other threads on here somewhere, so was my last hybrid and the one before that.

[rjent]

I wouldn't be too concerned about actual gallons filled as to what the computers are saying. It is next to impossible to get the tank to the exact fill each time, so it will skew your findings anyway. I always try to fill the tank to overflowing, and I still get very different values each time. I am not sure we can accurately fill the tanks the same every time. There are too many factors as to how it fills (angle, time of day or heat, ect)

 

JMHO

[Guardian_Bob]

I wouldn't be too concerned about actual gallons filled as to what the computers are saying. It is next to impossible to get the tank to the exact fill each time, so it will skew your findings anyway. I always try to fill the tank to overflowing, and I still get very different values each time. I am not sure we can accurately fill the tanks the same every time. There are too many factors as to how it fills (angle, time of day or heat, ect)

 

JMHO

 

Absolutely you're right that you cannot fill to the same level each time, which is why I was suggesting using 5 fill ups. Over time that error will fall into the noise range. For example, over 5 fill ups at 10 gallons a piece you'd expect to have the computer be at 50 gallons. So if the fill up is off by a half gallon, that'd result in 49.5 gallons, off by 1%. So you'd expect the computer and your notes to be pretty close at that point.

 

The goal is to check the actual input vs the amount the computer says came in.

[acdii]

Even at first click I always wind up with more gas in than the trip says I used. If you want real accuracy use a scan gaugeii. Once calibrated after 3-4 tanks each fill will be within .2 of a gallon. I used it on the first car, but its in the glove box of the Flex in case I need to tow something, so I can keep an eye on trans temps. I have no need to use it in this car, it does what I expect it to.

[Charrua]

After 10000 miles and 25 fuel-ups I can say with certainty that my FFH display is off by an average of 1.2 MPG. That's assuming Fuelly does their math right. Fuelly lifetime is 39.8, FFH lifetime (never reset) is 41.0.

[djminfll]

I hate filling the gas tank right to the top because invariably, no matter how careful I am, at some point it will overflow and drip down the rear fender. So I consistently fill the car until the first time it clicks off. I always assume that this happens at a consistent point. Is this incorrect?

[acdii]

I hate filling the gas tank right to the top because invariably, no matter how careful I am, at some point it will overflow and drip down the rear fender. So I consistently fill the car until the first time it clicks off. I always assume that this happens at a consistent point. Is this incorrect?

Sort of, depends on temperature and fill rate. If you fill at the same pump every time, and at the same temperature, then the probability of a consistent fill is very high.

[hybridbear]

It could be too that the gas pumps are inaccurate. I wouldn't doubt that even seemingly reputable gas stations have their pumps set to the maximum allowable discrepancy by law to cheat their customers. Even if the gas station only charges you for 1% more than you actually got that would be hundreds of dollars per day. And news stories have commonly exposed gas stations cheating customers by way more than 1%.

[acdii]

I know of one such station that has pumps that are inaccurate. I stopped to get diesel at this one station, and somehow managed to pump 7 gallons of diesel into a 6 gallon container.

[jeapa]

I had a little less then 1/8 of a tank so I filled up today. Only got a little over 9 gallons in. I pumped until 1rst click then put a little more in and stopped at second click. My fuelly milage was only .3 off from what the car said.

[Charrua]

It could be too that the gas pumps are inaccurate. I wouldn't doubt that even seemingly reputable gas stations have their pumps set to the maximum allowable discrepancy by law to cheat their customers. Even if the gas station only charges you for 1% more than you actually got that would be hundreds of dollars per day. And news stories have commonly exposed gas stations cheating customers by way more than 1%.

 

Interesting point. Someone on this forum must know how accurate a gas station pump must be to pass the inspection. It's hard to believe it would be 3%, which is the discrepancy I see.

[mrobinso8]

I kept a log of gas consumed over a 2500 mile period. The difference was 2 mpg lower than Ford's computer said. The same 2 mpg difference with my traded in 2011 Fusion. My 2006 Honda Accord would be off at the most .09 mpg, regardless of single tank computations of miles divided by gallons or a trip. This indicates that gas station meters are more accurate than many think. Could it be that Honda uses more accurate fuel use metering devices? Or, if you are a conspiracy theory type that Ford is doing it on purpose to make product look better that it currently is? My last tank car said 46.9 mpg and calculation was 44.7.

 

PS My Audi was spot on.

[acdii]

The department of weights and measures(I think thats what they are called) comes out with a special trailer that has accurate measuring equipment. They fill up a container with graduation marks on it, and compare it to the pumps. How accurate they need to be to get the certification stamp, I dont know, but thats how they do it. I dont know what they do with the gas after they did their tests though.

 

 

Hmm just did a check, first one I came across was Texas. Apparently all you have to do is fill out a form and send it in with a fee to be certified. http://www.texasagriculture.gov/RegulatoryPrograms/WeightsandMeasures/Devices.aspx Doesnt say that they have to come out and test.

Edited August 6, 2013 by acdii

[acdii]

Here's a nice article from 2011. http://www.wthr.com/story/16013215/pump-problems?clienttype=printable

[terryb]

Per Google search, Measurement tolerance in TX for new pumps is +/- 1/4% and on going operation tests require +/- 1/2 %.

[hybridbear]

I kept a log of gas consumed over a 2500 mile period. The difference was 2 mpg lower than Ford's computer said. The same 2 mpg difference with my traded in 2011 Fusion. My 2006 Honda Accord would be off at the most .09 mpg, regardless of single tank computations of miles divided by gallons or a trip. This indicates that gas station meters are more accurate than many think. Could it be that Honda uses more accurate fuel use metering devices? Or, if you are a conspiracy theory type that Ford is doing it on purpose to make product look better that it currently is? My last tank car said 46.9 mpg and calculation was 44.7.

 

PS My Audi was spot on.

Every vehicle is different. I never really paid attention in previous cars until our 2006 Honda Accord Hybrid. That car was off by a minimum of 3 MPG per tank, sometimes 5-6 MPG. It seems that independent tests have found Toyota hybrids to be among the least accurate hybrids in tracking their own fuel use.

[Guardian_Bob]

So I know all of you have been waiting for this, but I wanted to make sure my data was correct.

 

 

5% difference. Well that doesn't sound so bad, at least until you do the calculation. Average fill up is 10.366 gallons so I'm missing more than a half gallon every time I fill up.

[lolder]

My 2010 was 4% too high but the odometer was 2% low for a net of 2% or 1 mgg too high.

[hybridbear]

So I know all of you have been waiting for this, but I wanted to make sure my data was correct.

 

 

5% difference. Well that doesn't sound so bad, at least until you do the calculation. Average fill up is 10.366 gallons so I'm missing more than a half gallon every time I fill up.

Thanks for gathering the data! Have you checked your odometer accuracy?

 

My 2010 was 4% too high but the odometer was 2% low for a net of 2% or 1 mgg too high.

Like lolder in the 2010, I have found that our 2013 odometer is off by 2%. I haven't found as large a discrepancy as you have in the gallons from the pump versus from the car so our odometer discrepancy almost perfectly offsets the trip computer inaccuracy.

[Guardian_Bob]

Thanks for gathering the data! Have you checked your odometer accuracy?

 

Like lolder in the 2010, I have found that our 2013 odometer is off by 2%. I haven't found as large a discrepancy as you have in the gallons from the pump versus from the car so our odometer discrepancy almost perfectly offsets the trip computer inaccuracy.

 

Odometer, no. But the speedometer I have checked. It is dead on using my phone GPS as confirmation. (Cruise control set speed, confirmed with phone on many occasions.)

 

My trip this morning was (according to my phone, checked with Google maps) 26.3 miles. Assuming that's accurate, the computer claimed my trip was 25.9 miles (rounded down as the computer always does for the trip). I know some days it says 26.0, and others 25.9 (driving the same route) so I believe it should probably round up. 26.3-26=0.3. 0.3/26.3=0.0114 or 1.1% under. If instead we use 25.9, 26.3-25.9=0.4. 0.4/26.3=0.0152 or 1.5% under.

 

Given tire inflation (can change your speed by +/- 10%) this is fairly reasonable.

[Charrua]

Thanks for gathering the data! Have you checked your odometer accuracy?

Odometer accuracy does not really matter because both the spreadsheet and the FFH computer are using the same odometer miles.

Let's say that the odometer is off by a factor of 2. So instead of traveling 4400 miles you actually traveled 2200 miles. All that means is that instead of of 42 MPG (spreadsheet) and 44 MPG (FFH) it was 21 MPG and 22 MPG. The percent difference in fuel usage remains the same.

[hybridbear]

Odometer accuracy does not really matter because both the spreadsheet and the FFH computer are using the same odometer miles.

Let's say that the odometer is off by a factor of 2. So instead of traveling 4400 miles you actually traveled 2200 miles. All that means is that instead of of 42 MPG (spreadsheet) and 44 MPG (FFH) it was 21 MPG and 22 MPG. The percent difference in fuel usage remains the same.

If the odometer is inaccurate then you will have travelled more miles than the odometer reports. If his odometer is off by 2% then the miles aren't 4337.8 but are actually 4424.6. If you divide 4424.6 by the gas pump gallons reported (103.664) you end up with 42.68 MPG instead of 41.84. That means that the trip computer MPG inaccuracy is 3% instead of 5%.

[Charrua]

If the odometer is inaccurate then you will have travelled more miles than the odometer reports. If his odometer is off by 2% then the miles aren't 4337.8 but are actually 4424.6. If you divide 4424.6 by the gas pump gallons reported (103.664) you end up with 42.68 MPG instead of 41.84. That means that the trip computer MPG inaccuracy is 3% instead of 5%.

Yes, but the trip computer would also think that it had traveled 4424.6 miles instead of the actual 4337.8 miles and therefore it would have reported (4424.6 miles / 98.59 gallons =) 44.88 MPG and NOT 43.00 MPG. Still off by the exact same percentage.

 

It does not matter how inaccurate the odometer is, the MPG discrepancy, as a percentage will remain the same.

[hybridbear]

Yes, but the trip computer would also think that it had traveled 4424.6 miles instead of the actual 4337.8 miles and therefore it would have reported (4424.6 miles / 98.59 gallons =) 44.88 MPG and NOT 43.00 MPG. Still off by the exact same percentage.

 

It does not matter how inaccurate the odometer is, the MPG discrepancy, as a percentage will remain the same.

But it doesn't. You're missing the point. We all know that the car underreports how much fuel it has burned. Every car does this. But this car also underreports how many miles it has traveled. If you're going to correct the fuel use by calculating based on gallons at the pump, you should also correct the miles inaccuracy. It isn't fair to correct one inaccuracy but not the other in your calculations. Saying that the trip computer MPG is off by 5% is not accurate. Since MPG is a factor of two components (miles & gallons used) it isn't truthful to adjust the gallons for the inaccuracy of the dash display without adjusting the miles as well for this same purpose.