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

Updated MPG Display Bias Calculations

Discussion in 'Gen 2 Prius Fuel Economy' started by enerjazz, Mar 20, 2012.

  1. enerjazz

    enerjazz Energy+Jazz=EnerJazz

    Joined:
    Feb 18, 2004
    256
    144
    4
    Location:
    Fairview, TX
    Vehicle:
    2007 Prius
    Model:
    III
    I track every tank and record the car display, miles, and gallons input. I've tracked and charted the difference across 7.5 years and two Prius vehicles.
    Web site
    Large pdf chart
    [​IMG]

    The data shows the display indicates 1.4mpg bias better than the actual mileage. That's about a 2.6% error. The display is consistant, but has a bias. The miles/gallons calc will vary tank to tank due to the bladder.
     
    3 people like this.
  2. lamebums

    lamebums Member

    Joined:
    Jun 26, 2008
    101
    30
    0
    Location:
    Southern Ohio
    Vehicle:
    2007 Prius
    Model:
    N/A
    I'll have to start one of these myself. I fueled up on the 13th and the FCD claimed 57.7 MPG, and hand calc was 57.1, so it was more or less spot on.

    Just pulled in from the gas station 10 minutes ago with an FCD claiming 62.7 and the hand calc saying 46.73.

    Has anyone else seen a 20+% bias before?

    I -suspect- it's just the bladder acting up, because I don't think I've ever managed to cram 9.6 gallons in there when I still had two pips left.

    Either it's the bladder or the owner of the gas station has been pit-monkeying with the weights and measures.

    More to come, I'll start a log book and update as often as I can.
     
  3. enerjazz

    enerjazz Energy+Jazz=EnerJazz

    Joined:
    Feb 18, 2004
    256
    144
    4
    Location:
    Fairview, TX
    Vehicle:
    2007 Prius
    Model:
    III
    If you just got much warmer temperatures the bladder probably softened up and let you put much more gas in the tank. I often notice big swings in fall and spring as the weather changes and the bladder becomes more or less rigid.
     
  4. lamebums

    lamebums Member

    Joined:
    Jun 26, 2008
    101
    30
    0
    Location:
    Southern Ohio
    Vehicle:
    2007 Prius
    Model:
    N/A
    Hi enerjazz--


    Im thinking it may have just been a really, really high fill last time, because I've driven 210 miles on the new tank and still haven't lost a pip yet (when I normally lose one at about 100 miles). I'll keep you posted.
     
  5. jpadc

    jpadc Type before I think too often

    Joined:
    Sep 9, 2008
    343
    34
    0
    Location:
    Indianapolis and Northern Wisconsin
    Vehicle:
    2008 Prius
    Model:
    N/A
    Are you sure all that bias is in the display and none at the pumps? I suppose what matters most is what you are paying out of pocket, but the idea that gas pumps are accurate to the 1/1000 of a gallon (which is a little more than 1/2 a teaspoon is not very likely either...
     
  6. babybird

    babybird Member

    Joined:
    May 1, 2011
    117
    44
    0
    Location:
    Colorado
    Vehicle:
    2008 Prius
    Model:
    N/A
    In my old Subaru it was easy to track pump bias because in 5.5 years of driving that car my mileage varied by a whopping .8 MPG. So when I'd try a new gas station and my next tank was more than 1 MPG lower than normal doing exactly the same driving, it was pretty obvious what was going on. In my Prius, it's not so easy and requires several tanks to notice a difference for sure.

    I've seen large swings in mileage due to my bladder too. When I was averaging 53-57 MPG tank to tank, I suddenly got one that calculated out to 63.3 MPG on the same driving, and the next tank was 47.7 MPG after which it went back to my usual 53-57 MPG.
     
  7. Son of Gloin

    Son of Gloin Active Member

    Joined:
    Jun 16, 2008
    421
    320
    0
    Location:
    Lafayette, IN
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Premium
    After 104 fill-ups in our '08, I've had a 10%-or-more bias a total of only five times, and had one REALLY quirky one of about 34%; 73.6 mpg displayed vs. 54.8 mpg H/C! HOWEVER, over the long haul / 104 fill-ups, the overall bias is a bit less than 2%, with the H/C being the lower result, of course ( about 49.6 vs. 50.5 ) ...

    I wouldn't worry about it until after your next fill-up ... it probably IS a "bladder" issue, and it probably will balance out somewhat - if not entirely - after your next pit-stop.
     
  8. enerjazz

    enerjazz Energy+Jazz=EnerJazz

    Joined:
    Feb 18, 2004
    256
    144
    4
    Location:
    Fairview, TX
    Vehicle:
    2007 Prius
    Model:
    III
    There are many factors that cause subtle variations in mpg calcs - tires, worn tires, bladder, etc. I'm just noting that over 7 years there is a fairly consistent difference in my display data and my total miles driven divided by gallons purchased. I find the dashboard display to be more consistent (calc is always variable because of the bladder) - and it's accurate if I subtract 1.4mpg.
     
  9. lamebums

    lamebums Member

    Joined:
    Jun 26, 2008
    101
    30
    0
    Location:
    Southern Ohio
    Vehicle:
    2007 Prius
    Model:
    N/A
    Hi Son of Gloin--

    Stupid bladder, I think. I filled it up today, just after it hit 3 pips (Kroger/Shell discount expires at the end of the month...) and it clicked off at 5.9xx gallons. I'd driven 501 miles since the last fill-up, or an impossibly high 86.x MPG.

    A hundred clicks and some head scratching later, I was able to get 6.952 gallons in there - or 72.07 MPG. MFD claimed 61.4.
     
  10. RRxing

    RRxing Senior Member

    Joined:
    Jul 7, 2009
    2,516
    1,789
    0
    Location:
    NEPA
    Vehicle:
    Other Hybrid
    Model:
    Limited
    My average display/calculated variance since July 2009 is 5.1926%. Granted, this is for a 2010, but I thought I'd put it in this Gen II thread for reference. :)
     
  11. tdelker

    tdelker Junior Member

    Joined:
    Feb 6, 2012
    46
    4
    0
    Location:
    Colorado
    Vehicle:
    2004 Prius
    Model:
    II
    I'm curious about this. I understood the the mpg for my subaru and all OBDII vehicals is calculated using injector pulse width (on-time)*known fuel flow through injector per unit time*number of on times/miles.

    Given that, the only way it could be off is if the known fuel flow through the injector or the miles traveled is wrong. Since the miles traveled is common in both the hand calc and the computer calc, then it can only be differences in fuel flow through the injectors.

    Or does the Prius not do the calculation that way?

    Thanks,

    Tom
     
  12. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,215
    1,200
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    Yes but there's always going to be a percentage error in the "known" injector flow rate as well as in the time intervals. You cant expect 0% error in a system like that.

    By way of analogy, say you took a pedometer and calibrated it with your "known" step length and then walk around for a week with it. Do you really think the measured distance traveled would be anything other than an approximation?
     
  13. lamebums

    lamebums Member

    Joined:
    Jun 26, 2008
    101
    30
    0
    Location:
    Southern Ohio
    Vehicle:
    2007 Prius
    Model:
    N/A
    Hi enerjazz--

    I have a larger sample size by now and will update as I update my logs. It's 3+ AM where I am but some rough calc's I have indicate the FCD is about 6% optimstic. I had to throw out a couple sub-6 gallon fills as sometimes I'd top the car off on a hunch that gas prices would spike soon (or had already spiked at some stations and not at others yet). I'll get the precise calculations to you ASAP.

    Now if only I could get a tank of gas to last me a full month, so I could fully capitalize on Kroger's 10 cents off a gallon deal. It's distressing because the longest I've managed to make a tank stretch is twelve days.
     
  14. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,215
    1,200
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    My calculated consumption is normally about 0.1 L/100km over the MFD displayed value. This is typically 0.1 in 4.2, so about 2% to 2.5%

    It's interesting to note that the L/100km display that we have (Australia Canada UK etc) has far less resolution in the high MPG region, compared to the US MPG display. For example, the difference between 58.0 and 58.1 MPG is only 0.007 L/100km, which corresponds to over 14 times higher resolution than the 0.1L/100km limit that we have! Obviously this makes accurate calculation of this bias far more difficult for we L/100km guys.
     
  15. RikM

    RikM Junior Member

    Joined:
    Jan 15, 2012
    35
    3
    0
    Location:
    UK
    Vehicle:
    2008 Prius
    Model:
    II
    Er, we don't get l/100 km in the UK.... We have MPG, but imperial gallons...
    20% bigger than the US. So an even coarser scale.

    I've seen some calls recently for a switch to "miles per litre", as the pumps don't dispense in gallons anymore. The kind of awkward mismatched mess we've been used to in the last 30 years or so. At least I'm too young to remember using "old money"...
     
  16. uart

    uart Senior Member

    Joined:
    Sep 7, 2009
    4,215
    1,200
    0
    Location:
    Australia
    Vehicle:
    2005 Prius
    Whoops, I got that mixed up (most of the other regional differences are similar in the UK and Australian models). Anyway, L/100km is standard for Australia, Canada and much of Europe. Apparently Japan and some others use km/L.

    .
    That's an interesting mix Rik, so all your petrol pumps are now in litres. Do most people still "think" in terms of gallons or have they mostly switched to litres?
     
  17. RikM

    RikM Junior Member

    Joined:
    Jan 15, 2012
    35
    3
    0
    Location:
    UK
    Vehicle:
    2008 Prius
    Model:
    II
    I think it may be a generational thing. Most of the younger drivers have grown up with cm, litres, grammes etc. but also miles (no switch to km planned) and mph.
    I know my 12yo has trouble visualizing feet and inches.
    Much the same way that I can't quite picture how people lived with 240 pence to a pound...

    But I'm agnostic about imperial/metric measurements, as I grew up with both, with maybe some bias towards metric as it was the rule in school science.

    Ask me for a distance to a place, and I'll most likely answer in miles or yards. Ask me for the width of something small and I'll use mills. In between I might use cm or inches, depending. Ask me for cooking instructions and I work in grammes. Ask me my weight and if I answer it would be in stones and pounds...
    :)