1. Attachments are working again! Check out this thread for more details and to report any other bugs.

Accuracy of consumption calculations

Discussion in 'Gen 3 Prius Main Forum' started by gconrad, Mar 25, 2013.

  1. gconrad

    gconrad Blizzard Brigade #221, the FIRST Persona...

    Joined:
    Mar 2, 2013
    19
    6
    0
    Location:
    Virginia
    Vehicle:
    2013 Prius
    Model:
    Persona
    I have searched and read many threads concerning this... The consensus is that the MPG calculated by the Prius tends to be optimistic by a varying degree, typically 1-2MPG over actual consumption.

    On my first fuel-up I had ~20 miles cruising range remaining, 472 miles on the clock, and put 8.962 gallons in for a manually calculated 52.66MPG. The consumption computer showed 47.8MPG.

    Today I filled up for my second time. 8 miles cruising range remaining, 905 on the clock (433 trip), put 9.511 gallons in for a manually calculated 45.52MPG. The consumption computer showed 48.1MPG.

    My two questions are this:

    1. Is the computer off THAT much for the typical owner? (-4.6 to +2.6) Doesn't seem to be the case from the threads I have read.

    2. Does the fuel shut-off hit early on the Prius for some reason and require a "top-off"? I have always stopped pumping fuel when the pump stops. Fuel gauge shows full after filling but with ~20 miles remaining I should have put much more than 9 gallons in and then today with 8 miles remaining I should have put much more than 9.5 gallons in. This to me is just confirmation that the consumption computer calculations are off, heavily if I may so...

    Thoughts?
     
    #1 gconrad, Mar 25, 2013
  2. xpcman

    xpcman Senior Member

    Joined:
    Jun 11, 2009
    1,302
    295
    0
    Location:
    California - SF Bay area
    Vehicle:
    2008 Prius
    In general you can't do an accurate MPG calculation when you use a standard pump. Also you rounded the miles traveled to zero decimal places then used 3 decimal places for your gas consumption.

    Paragraph T.2.1.2 (b) of the National Institute of Standards and Technology Handbook 44, “Devices Indicating in Inch-Pound Units,” says that:
    The acceptance tolerance on normal and special tests shall be 1/2 in3 plus 1/2 in3 per indicated gallon and never less than 1 in3.​
    This means that the accepted inaccuracy of a gas pump is half a cubic inch (over half a US tablespoon) plus half a cubic inch for each gallon and never less than one cubic inch. When you buy ten gallons, the pump can over- or under-dispense five and a half cubic inches of gas and still be in compliance.
    So how much is five and a half cubic inches of gas? About 0.024 US gallons, or over twenty times the accuracy implied by showing a thousandth of a gallon on the pump. On one gallon of gas or less, the allowed inaccuracy is 0.004 gallons, or four times the implied accuracy.
    This is also the best case scenario: Paragraph (a) doubles the tolerances for machines between repair intervals. Heat can also play a significant role, as all industry tolerances assume 60°F — a gallon of gasoline at 100°F takes up almost 3% more space.
     
    #2 xpcman, Mar 25, 2013
  3. gconrad

    gconrad Blizzard Brigade #221, the FIRST Persona...

    Joined:
    Mar 2, 2013
    19
    6
    0
    Location:
    Virginia
    Vehicle:
    2013 Prius
    Model:
    Persona
    While I understand the inaccuracies given the trip odometer only goes to 1/10th and the pump calibration allows for error, these are tiny in comparison to the Delta seen on the consumption computer.

    To demonstrate:

    Utilizing this information and fill-up #2 from today... Assuming worst case scenario where I rounded down by .5 miles on the trip and the pump was as inaccurate as NIST allows between service intervals at .48% (.024 gallons per 10 doubled) and therefore only pumped 9.465 gallons, the calculated MPG would then be 45.8MPG (.28MPG better than my original calculation) and still a Delta of +2.3. Applying the same principle to fill-up #1 would net 52.36MPG (.30 MPG worse than my original calculations) and still a Delta of ~4.6.

    Weather/heat is not a factor as our temps have been around 50 on each fill-up.

    I understand that the computer nor manual methods will be dead on but I did not expect to see an error range of nearly 7MPG... Pretty much makes the computer a non-reliable novelty unless it calibrates itself over time and therefore becomes more accurate.
     
    #3 gconrad, Mar 25, 2013
  4. Munpot42

    Munpot42 Senior Member

    Joined:
    Mar 12, 2012
    1,391
    543
    0
    Location:
    Santa Monica, Ca. 90405
    Vehicle:
    Other Hybrid
    Is Your 1st tank calculation based on a dealer fillup, or did you fill it up yourself, makes a difference. I think you have too few fillups to guage the accuracy of anything, wait until about 10 fillups, you will have a better idea of where you are at. The miles remaining indicator means nothing in your calculations, I doubt it has any accurate use as other than a very general guide as to when you might run of of gas.
     
    #4 Munpot42, Mar 25, 2013
  5. fuzzy1

    fuzzy1 Senior Member

    Joined:
    Feb 26, 2009
    17,105
    10,038
    90
    Location:
    Western Washington
    Vehicle:
    Other Hybrid
    Model:
    N/A
    Two thoughts:

    (1) For fuel tracking, don't get concerned about any fuel or mpg calculations for miles traveled before the first time you personally filled the tank. The dealer fillup doesn't count because you don't know just how they filled it, nor when they reset the trip meter. The numbers here suggest that the dealer overfilled it, or didn't zero the trip meter.

    (2) The 'distance to empty' is not until it is really empty. It includes a considerable safety margin, just as the low fuel lights and 'E' marks on most other cars also contain some safety margin. Fuel consumption varies enormously with weather, driver style, mountain grades, and long vs. short trips, and the car cannot accurately predict what lies ahead. So it leaves a considerable safety margin so Customer Support doesn't get lots of angry calls from irate customers who ran out of gas early despite showing 100 miles remaining as they passed one of these signs:
    [​IMG][​IMG]

    For an idea of how far a typical Prius might go after DTE = 0 miles, read the first post of this thread: [WARNING] Running out of gas (Gen III). But remember that gauging also varies between cars, and some owners have to draw the short straws. That could be you.

    (FYI, The real story behind the car's out-of-gas behavior is discovered very deep in the thread.)
     
    #5 fuzzy1, Mar 26, 2013
  6. 3PriusMike

    3PriusMike Prius owner since 2000, Tesla M3 2018

    Joined:
    Jun 21, 2009
    2,938
    2,288
    0
    Location:
    Silicon Valley
    Vehicle:
    2012 Prius Plug-in
    Model:
    Plug-in Base
    You can easily refill a few tenths of a gallon different each time you fill up. So you need several fill ups to minimize the possible error and average over all fill ups.

    Mike
     
    #6 3PriusMike, Mar 26, 2013
  7. gconrad

    gconrad Blizzard Brigade #221, the FIRST Persona...

    Joined:
    Mar 2, 2013
    19
    6
    0
    Location:
    Virginia
    Vehicle:
    2013 Prius
    Model:
    Persona
    Thanks for the replies... Mods, this wasn't meant to be as much a fuel-economy discussion as it was a discussion regarding the computer's accuracy, feel free to move to fuel-economy forum if need be!

    To clarify a few things...

    1. I understand that the DTE has nothing to do with the calculations, I only included that to show that I had 8 miles DTE and still only put 9.511 gallons in, leaving more than 2 gallons in the tank and therefore ~100 miles DTE. I consider the last PIP blinking the low-fuel warning light and the DTE a nice to have calculation.

    2. The car had 1 mile on it when I took delivery. It had never been test driven, I took delivery off the truck. Therefore, worst case scenario if the dealer did not reset the trip meter it would have been 471 vice 472 making only 1/10th of a mile difference in the manually calculated MPG. I also understand that the dealer could have filled it up and topped it off until spilling out of the filler neck, a difference of a few tenths to maybe even half a gallon. Still, adding in a half-gallon of gas, the manual calculation would have been +2.8 over consumption computer.

    The variations in all of the calculations are understood, and the inacurracies of trip meter/pump method are understood as well. I don't expect anything to be perfect, I was simply surprised that the consumption computer was off by as much as 10% +/-. All of the mentioned errors, miscalculations, and theories added together equal a couple % at best...

    My responses are meant to argue the validity of anyone's input, rather to add to the discussion. Now I'll quit nuking it and worrying with it until I get to 10 fill-ups at which point I will re-evaluate. Thanks everyone!
     
    #7 gconrad, Mar 26, 2013
  8. CaliforniaBear

    CaliforniaBear Clearwater Blue Metallic

    Joined:
    Oct 9, 2012
    1,179
    289
    0
    Location:
    Northern California
    Vehicle:
    2012 Prius
    Model:
    Plug-in Base
    In my 2012 Prius Three with 12 fill-ups the indicated consumption averaged 3.5 mpg higher than measured gas into the tank and odometer miles. The odometer was very accurate, speedometer not so much. Things are more complicated now that I have a Plug-in :)
     
    #8 CaliforniaBear, Mar 26, 2013
  9. fuzzy1

    fuzzy1 Senior Member

    Joined:
    Feb 26, 2009
    17,105
    10,038
    90
    Location:
    Western Washington
    Vehicle:
    Other Hybrid
    Model:
    N/A
    No, it (or at least some of them) can take more than that. I believe several readers here have put in well over a whole gallon extra. Mine has never been filled to the filler neck, but a few times has been loaded with more than a half gallon extra after the first trip.
     
    #9 fuzzy1, Mar 26, 2013
  10. gconrad

    gconrad Blizzard Brigade #221, the FIRST Persona...

    Joined:
    Mar 2, 2013
    19
    6
    0
    Location:
    Virginia
    Vehicle:
    2013 Prius
    Model:
    Persona
    Good to know, hence the reasoning behind question 2 in the original post. I am not used to a vehicle that takes that much more after the auto shut-off. On my other vehicle if you pump more than a very small amount (2/10ths maybe) after auto shut-off you will be overflowing at the filler neck. Thanks...
     
    #10 gconrad, Mar 26, 2013
  11. HprDad

    HprDad Junior Member

    Joined:
    Sep 25, 2012
    72
    17
    3
    Location:
    WI
    Vehicle:
    2012 Prius
    Model:
    Four
    Over 17 fill-ups, mine reads high anywhere between 2.5 and 3.5, compared to pump calcs.
     
    #11 HprDad, Apr 10, 2013
  12. Xyrus

    Xyrus New Member

    Joined:
    Apr 10, 2013
    9
    5
    0
    Vehicle:
    2013 Prius
    Model:
    Two
    You also need to take into consideration fuel formulation, which can vary. The consumption calculator doesn't know whether or not you put in E10 or pure gasoline or a gasoline mixture with some other additives. It also doesn't take into consideration temperature expansion or contraction of the tank/fuel, evaporation, or any other elements that can affect MPG calculations. That's why even manual calculations will be off (though not as much as the computer with fixed parameters).
     
    #12 Xyrus, Apr 10, 2013
(You must log in or sign up to post here.)