computer mileage different then reality

I would say get over it.... your error means that over the 1600 miles there was a discrepancy of 2.5 gallons of consumption.

That's 2.5 gallons out of ~ 38 gallons, or 6.7%. So, how full was the tank when the test started? How full was it when the test ended?

How accurate is the odometer? A 2% error here would mean + / - 32 miles... or almost a gallon of fuel.

Also, how accurate is the fuel metering system used for the mpg calculation in the car?

I think the better way to look at it is that the MPG indicator gives you a relative sense of performance and not an absolute measurement. So, watch for changes in its readings... and then, if you want, just deduct a couple of mpgs from it... and just know that will be pretty close to what you are actually getting.

 

Hey, another AR!

Here's a way to channel your anal tendencies into a positive endeavor: Start tracking on a spreadsheet the differences between MFD MPG and calculated MPG (actual miles driven divided by gallons filled). Before drawing any conclusions, continue it over several tanks in a variety of weather conditions.

I started doing this myself in October, and as of the latest fillup I show the MFD to overestimate fuel economy by 1.2%. Some here have tracked the difference for a year or more, and also report a fairly small overall difference.

There really is nothing to "fix." The car's computer calculations of fuel mileage are close but not exact. So your other option, as you suggest, is to get over it.

 

This motivated me to upload my spreadsheet for anyone to use, posted here.

 

You can't assume that gas pumps are more accurate than the MFD. The law allows a few percent of error, and gas station owners have a financial incentive to have their pumps deliver less gas than is displayed on the pump. And that makes manually calculated MPG less than the true MPG.

 

Hello,

MY 2 cents: I've got the 2008 Honda Civic Hybrid and it has, over the three tanks I've had it, UNDERESTIMATED my mileage by a few mpg so I think the onboard computer calculations are by nature imperfect. My car's MPG says 43.5 when I'm calculating 47.2 which for this very cold Chicago winter is fantastic. My neighbor owns a 2005 prius with 38k miles now and gets 46mpg according to the display (lifetime) but he's never hand-calculated it like I'm doing but I'm going to ask him to do it once or twice to see what the descrepency is as far as mpg's are concerned. As far as I'm concerned calculating by hand is always going to give you the most precise numbers.

 

The MFD should be pretty accurate. I believe it measures the fuel flow at the injectors. To my knowledge, it is very accurate. In the early days of owning my Prius, I liked to push the limits of the tank. I once filled the tank to it's capacity, and tried for 1000km for that fill. My MFD reading was bouncing between 4.4L/100km and 4.5L/100km near the end of the tank. After the trip meter rolled to 1000km, the ICE sputtered with it's last sips of gas. The final MFD reading was 4.5L/100km over 1001km. That was pretty dead-on, if you ask me.

 

It's not a few percent. The law allows one cubic inch absolute plus one cubic inch per gallon. A gallon is 231 cubic inches. So for a typical eight gallon fill of a Prius, that's about half a percent allowed error. Government inspections assure that this is not exceeded by much.

Half a percent is better than I would expect a fuel flow measurement to be (I'd guess a few percent for that), so you should be able to treat the reported amount of gas pumped as a better measure than the total gas injected into the engine according to the Prius computer.

Assuming that all the gas pumped ends up in the tank of course -- I have twice accidentally overfilled with gas on my shoes.

There are also small variations in the density of gas as a function of temperature. Though the volume is still the volume, what we're really interested in is the energy in the gas, which is a function of the mass. Having said that, gas stored in underground tanks stays at a pretty constant temperature, show it shouldn't make much difference.

 

And... for those many people who can never really "fill" the tank due to premature pump shutoff (we need a product like Prialis), the hand calculation is far less accurate. My hand calculations show 42 mpg, but the MFD reports 37.

Since I reach 2 pips at < 290 miles, I'm inclined to believe the MFD minus 1 or 2 MPG's.

 

I've been doing this for about 9 months now and my avg. difference is around -2. I'm glad somebody started this thread and that I'm not the only one seeing this. I'm still happy with the mileage I'm getting but this discovery was indeed a letdown. I would expect a small degree of difference either way but if it is consistently estimating avg FE to be over by 2 mpg on most cars I will have a hard time being convinced the MFDreading is not intentionally padded by Toyota.

 

My other car (not a Toyota)... displays average MPG... and its tank is very easy to top up. At 40,000 miles' worth of fill ups, the reading on the display is consistently 4 to 6 % higher than actual... with a display of 25 MPG, I am actually getting slightly less than 24 MPG.

I do think there is an upward bias in the displayed number in all of these cars. But I don't think it is done for a mischievious purpose...

For example, one thing I am aware of is that for safety reasons the speedometers are biased to read high. Since the odometer is integral to that reading, you automatically are going to get higher MPG displayed than actual.

Another issue - related to the speedometer: As your tires wear out, they are actually getting smaller in diameter and spinning faster for the same speed... resulting, again, in a higher speedometer reading then when the tires are new. (the difference in diameter between a new and worn tire is about 2%)

I don't know the type of fuel meter used, but if it is positive displacement or even a propeller type - it will have some slip (where fuel gets past the meter without the meter reading it)... This will tend to report less fuel used then actual... and, again, result in a better than actual MPG being displayed.

Its an interesting problem, and I would not mind it being more precise... but I realize limitations are going to exist, especially since Toyota would only want to spend very little money on this type of instrumentation. So the issue is accuracy at what price... and given the purpose... my feeling is that a 5% accuracy for this application is very adequate.

And... so... I (we) should just live with it... just be aware of it and account for the discrepancy in your head.

Speaking of your head (the ultimate computer)... have you noticed a bias in your own reporting of MPG in the Prius to friends... You know: remembering the better numbers, putting the car in a favorable light.... its just human nature...

 

There are really a lot of issues here:

The gas pump may be off, and it's in the financial interest of the gas station to short you.

The car's calculation may be off, and it's in the interest of the car company to make you think your mileage is better than it really is.

Tire size and speedometer/odometer inaccuracy do not affect the comparison between calculated reading and car computer report, beause you're using the same odometer in both cases, but if you measure your odometer inaccuracy you can take that into account for a more realistic number.

With the Prius more than with other cars, tank to tank fill differences mean that you must average over a very large number of fill-ups, because the fill level will probably be different from one fill-up to the next.

For myself, I couldn't care less!!! The car is the cleanest and most efficient gas-burning car you can buy today, and of all gas-burning cars on the road, only the two-seater Honda Insight gets better mileage. While the kind of detailed records kept and published by john1701a are very informative and useful for people researching cars, there are so many factors that go into mileage that my actual mileage is irrelevant, and knowing my actual mileage will not change it. The MFD gives me useful relative mileage information for comparing different kinds of driving (routes, driving style, etc.) and that's all I need. Anything else is just a kind of video game, or a bragging contest.

 

i'm on the same boat on this one, i'm happily logging all my tanks in blissful ignorance of the variance. it makes me feel better. <whilstling while skipping>

 

Reality is subjective. Relax.

 

My bad, I should have written that some states allow up to a few percent error. Good for CA in requiring better accuracy. Also note that no US gas pumps are temperature compensated, unlike pumps in Canada. And of course some gas retailers are pretty clever about cheating, for example setting their pumps to give an accurate reading when the amount used by the weights and measures folks is dispensed but cheat otherwise (this was used in NJ some years ago).

 

In some locations there are no inspectors to check the pumps, so it doesn't really matter what the law says.

As for the Prius fuel metering, it is done by the injectors. The engine controller needs very good information about how much fuel is being injected to keep the fuel ratio at the correct level. If the injector calculations are off, so is your mixture.

This was stated briefly above, but keep in mind that the usable capacity of the Prius fuel tank in the U.S. model varies considerably. There is no way to know how much fuel was used in a tank for doing manual calculations. Manual mileage calculations for the U.S. Prius are only meaningful over a large number of tanks.

Tom

 

I must chime in. I'm a new owner. I just passed my third tank of gas.

I reset the MPG after my last fill. Computer tells me 48.6 MPG when I pull into the gas station.

I filled it (and let it stop when the pump stopped naturally. No overfilling)

472 miles and it took 10.38 gallons to fill it. Ergo - 45.45 MPG in actually.

Difference? 3%. Not huge, mind you, but it makes me wonder about whether Toyota is purposely overstating what you're getting.

On another note, since I filled my tires to 44/42 (F/R), my mileage has ticked up 2.5 MPG.

 

for some reason, it won't let me save changes to my message.

anyway, it's 6% not 3% difference. That's quite a bit.

 

I once pulled up to a petrol pump and lifted the nozzle but before I got the nozzle in the filler neck the pump had registered 70 cents worth of fuel. I looked for it on the ground and it wasn't there so I went and told the attendant and I moved to the next pump. The second pump was much better, only 30 cents worth before a drop flowed. I told the attendant (who I believed was the owner) to F**k off and went else where to get fuel.

Always look before you put the spout in the hole.

Oh yeah, speedo inaccuracy isn't always going to mean the odometers is wrong. My bike has 11% speedo error but a 1.5% odometer error.