Fraud By Algorithm?

Discussion in 'Clarity' started by Fast Eddie B, Mar 6, 2019.

To remove this ad click here.

  1. Mariner91

    Mariner91 Member

    No. Because that 199.99 will still be weighted by all the miles you were on EV previously. you get "true MPG" by resetting either trip A or trip B and driving only on HV. I just finished a 2 week trial of just being on HV, and that 199.99 went down to 160 after 2 full tanks of gas. Actual MPG (on Trip B that I reset on gassing up) was 52 MPG after about 600 miles driven on HV (multiple trips, fully charging every night). Trip B on single long trips on freeway is closer to 47-48 MPG.
     
  2. To remove this ad click here.

  3. MPower

    MPower Well-Known Member

    My 2012 Prius Plugin was the first car I had that estimated remaining mileage. The reality check was nearly being stranded in a snowstorm with no gas.

    The algorithm seems to think that we are all travelling on a smooth flat road, with one very svelt person driving very slowly in broad daylight with no heat and no luggage.

    I don't pay any attention to the gauges other than the gas and SoC bars. I do look at the estimated mileage everyday just to get a feel for how the temperature change are affecting the estimate.
     
  4. Candice

    Candice Active Member

    Forgive me for dumbing this way down. I am not well versed in car lingo or mechanics.

    If I jump into the Clarity today and it states I can go 320 miles on gas alone and I hyper-mile as much as possible for a long trip and am able to go 330 miles, is the algorithm on the car wrong?
    Same scenario in reverse...if I start with 320 mile range and drive extremely inefficiently with aggressive accelerating and decelerating and only get 310 miles, is the algorithm wrong?
    I guess my question is, how does the car know the type of driving you are going to do? Maybe it is stop and go traffic or windows down etc that affects the final number.
     
    Mariner91 likes this.
  5. 4sallypat

    4sallypat Active Member

    All cars I have had never came close to match the exact EPA MPG.
    Every Japanese, American, German, British car I have owned have had this issue.

    2 extreme cases I have personally seen:
    1. 2011 BMW 335d (diesel) EPA was sold as 35MPG on the highway but I got over 50MPG every single time I got on the freeways.
    2. 2016 Land Rover gas EPA is listed as 26MPG on the highway but I get only 22MPG every time I am on the freeways only.

    These are calculated by hand - tank to full, miles driven.

    Should I claim fraud for all the cars I have driven ???
     
  6. Dante

    Dante Member

    .... how does the car know the type of driving you are going to do? [/QUOTE]

    It doesn't, hence the EV range shows 47miles, but you can only go for 30 in a cold winter day, or HV range shows 300 for a full tank, but if you gun it it'll only carry you for 280.

    I'm a firm believer that the car's software (and not just Clarity but all, including any item that uses power ie. cellphones) calculate best guess and hope that you are as consistent in your driving patterns as possible... that would make the accuracy go up.
     
    Candice likes this.
  7. To remove this ad click here.

  8. craze1cars

    craze1cars Well-Known Member

    He stopped to fill up the tank 7 times, shutting the car off. And presumably stopped a few more times than that also for meals or bathroom or whatever. In my experience the car loses a couple miles of EV range every single time you activate HV mode, before it finds a new "level" point and maintains that level of charge. And you do have to reactivate HV mode every single time you shut the car off and turn it back on, because it always defaults to EV mode with a restart.

    So his gradual drop in EV range over a trip of that length does not surprise me one bit. Not necessarily because it's right, but simply because it is what my Clarity does on similar extended road trips as well.
     
  9. Richard_arch74

    Richard_arch74 Active Member

    I concur with @craze1cars. My experience from Mi to Tn all in HV mode: virtually everytime I restarted the car I would use a few electrons. When you think about it, if you go 2-3 hours between a rest stop, or gas fill up, or a meal you always have a new set point for your EV SOC. If you lose even just one mile of EV range per startup you may lose 5 miles of EV range per day. On my trip to Tn and back I lost about 12 miles of EV range while in HV the whole way.

    Sent from my SM-G955U using Inside EVs mobile app
     
  10. Wait.

    I was not referring to any predictive algorithm, which clearly will have all sorts of variables and uncertainties involved.

    But this mileage algorithm is a post facto calculation - no prediction or variables or guessing involved.

    For example, imagine you drove 200 miles and then it takes 5 gals to fill up. Even quick “mental math” shows you got 40 mpg. If the display shows 45 mpg, something is wrong with the algorithm. If it varied on both sides of the real number, one might assume imprecision. But when it’s virtually always erring on the high side, it certainly seems intentional.
     
  11. jdonalds

    jdonalds Well-Known Member

    What I've noticed is if the ICE has to work hard accelerating or climbing it will tap some power from the battery. On the other hand driving on a flat straight road with a steady speed will not use any battery power.

    I've driven 150 miles without the battery level dropping at all, even when it is charged well above the 58% mark.
     
    MPower likes this.
  12. To remove this ad click here.

  13. Mariner91

    Mariner91 Member

    Suggest switching Info view to Energy so you can see when it's all blue all around, including power coming From battery And Gas engine. HV mode is Supposed to use your battery power. Battery level does not drop because your gas engine goes back and forth on re-charging it
     
  14. Mark W

    Mark W Active Member

    CT
    I agree. I think we got off track with the range discussion. This is simply a discussion of the stated mpg for miles ALREADY DRIVEN. I'm guessing that the reason the calculation is inaccurate is that is uses a historical summation of the current instant mpg figures, which are not necessarily accurate. But that's just a guess.
     
  15. Thanks to Thomas Mitchell for that link.

    For those who did not click on it, in part...

    Roger Clark, senior manager of GM's energy center, explains that the fuel economy gauge makes a calculation by counting the number and duration of pulses made by the fuel injectors as they squirt gasoline into the combustion chambers of the engine. The onboard computer system divides the distance the car travels by this estimated fuel consumption.

    Clark says the gauge is "dead nuts accurate" — if you consider all the variables at work during driving, including temperature, driving conditions and driving style. The biggest fluctuation occurs because ethanol, which is blended with gasoline in varying amounts, contains less energy.

    "When you fill up, you are paying for a gallon of gas, but the energy in that gas varies significantly," Clark says. This means that while the car's computer assumes the gasoline is providing energy to drive a certain distance, the fuel might have less energy and not propel the car as far.

    The 5.5 percent average variation in the vehicles Edmunds tested "seems like a perfectly reasonable range to me," says Paul Williamsen, national manager of the Lexus College, where his responsibilities include service training for Lexus staff, dealers and corporate personnel. "I can't imagine any reason that any automaker would want to make drivers think they can get better fuel economy than they were getting," Williamsen adds.

    I don't buy "I can't imagine any reason that any automaker would want to make drivers think they can get better fuel economy than they were getting." If anything, that just shows a marked lack of imagination!

    I also don't buy the "ethanol excuse". Certainly, ethanol has an adverse effect on mileage, but the calculation of how many miles you went on a given gallon of E10 fuel or whatever would still be straightforward.
     
  16. So have I. As I've mentioned here or elsewhere, on 100 mile trips across the mountains I've gone to HV with 20 EV miles remaining, and in spite of excursions down to maybe 16 EV miles remaining on long uphill grades, it almost always drags itself back up to the "set" value eventually.

    But this was a 1,600+ mile trip, and it did not hold. Here's the details:

    [​IMG]

    As an aside, I would love a way to set on the display the mileage you want to maintain, and have the car do a better job of maintaining that level, even if it had to dip into "HV CHARGE" mode to do so.
     
  17. Walt R

    Walt R Active Member

    This may be unlikely, but would be easy to check. Does the reported MPG bias exist the same if the car reports in metric units? Maybe the Japanese engineers are just using poor conversion factors. Also, a UK imperial gallon is about 1.2 US gallons - is it too cynical to think they might have just split the difference rather than make a UK/Aus specific factor?
     
  18. 4sallypat

    4sallypat Active Member

    ^^^ very good point - the conversion of liters to gallons to imperial is not exact and the software rounds up/down the conversion ??
     
  19. jdonalds

    jdonalds Well-Known Member

    Right. Also in our two Prius cars we always got better mileage on the highway than we did in town. The EPA numbers were the reverse. I believe it all had to do with the fact that we aren't driving a steady 35 mph in town. Rather we are experiencing stop and go for stop signs, traffic signals, and traffic. Accelerating is the least efficient part of driving a car.
     
  20. Walt R

    Walt R Active Member

    To be pedantic, stopping is the least efficient part of driving a car. Accelerating doesn't lose energy, it just turns it into kinetic energy still embodied in the car.
     
    David Towle likes this.
  21. jdonalds

    jdonalds Well-Known Member

    Either way it is the lack of steady speed. Stopping and accelerating is contrary to good gas or electric efficiency.

    On top of that it really irks me when I stop at a traffic signal in our V8 4Runner with the engine running. Ugh.
     
  22. Ray B

    Ray B Active Member

    Do you have the OEM tires, or a different set? Any wear at all? How about tire pressure? Just checking as a different circumference can add to the distance calculation error.

    Not trying to poke holes in your data and conclusions, as I have the same suspicion about the mpg calc, which I have never found to be accurate in any car. Just trying to be thorough.
     
  23. Ray B

    Ray B Active Member

    Also, you can think about the error that is present in the gas pumps that could be measuring a different volume than is measured/tracked by the injectors and fuel system ECU. I think the pumps tend to be accurate as they get audited or calibrated regularly but always with reference to a specific temperature which may be quite different than when it goes into your tank, or when it is used by the car.

    Again, the car is likely the problem, but one should consider all the sources of error that can arise, some of which could be outside of the mpg calculation algorithm.
     

Share This Page