TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Can Never Match Trip Planner Prediction

Discussion in 'Model S: Driving Dynamics' started by mknox, Sep 16, 2015.

  1. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,565
    Location:
    Toronto, ON
    I've mentioned this in passing in several threads, but for some reason, I can never seem to match the Trip Planner prediction in the Tesla Energy app. I am always below the grey prediction line, and usually by quite a bit.

    Yesterday I had an approximately 25 mile trip to make after work and thought I'd try my best to meet or beat the prediction. It was over moderately hilly terrain (for Ontario), perfect weather in the 70s and I drove at or below the speed limit as gingerly as I could.

    I unplugged from my charger at work at about 92% SOC and as you can see below, my trip (Since Last Charge) came in at 255 Wh/mi which I believe is better than Rated. But I'm still below the prediction line on the Energy Trip app! This is about as close as I've come to it, however.

    What on earth do you have to do to meet Tesla's prediction?!?

    IMG_0456.JPG IMG_0455.JPG
     
  2. Zextraterrestrial

    Joined:
    Mar 11, 2010
    Messages:
    3,636
    Location:
    Humboldt/Los Altos
    strange.
    and strange the the predicted like vs the green 'actual' line have such different undulations. my trip lines have looked much more similar for the most part
    I would have probably ended at 85-6% or better with that trip.
    maybe there is a googly error in Miles to Kilometer conversion there? probably not
     
  3. tezzla

    tezzla Member

    Joined:
    Jul 14, 2013
    Messages:
    545
    Location:
    SoCal
    That is strange to me because I ALWAYS beat the estimate (averaging around 300 Wh/mi).
     
  4. Max*

    Max* Autopilot != Autonomous

    Joined:
    Apr 8, 2015
    Messages:
    4,871
    Location:
    NoVa
    I always beat the predictions too on longish drives.
     
  5. RiverBrick

    RiverBrick Active Member

    Joined:
    Mar 23, 2014
    Messages:
    1,370
    Location:
    Québec
    The pavement may be rougher in Ontario than in California. fwiw, I don't have much of a problem meeting or beating the estimate in the Summer. Winter is another story, of course. I wonder if the algorithm has been overhauled for severe cold.
     
  6. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,565
    Location:
    Toronto, ON
    This route was on smooth fairly recent pavement, so that's not the issue here. But even if it was, how do I manage 255 Wh/mi but still come in under the prediction? That's what has me stumped. Sure, if I was coming in at 350 or 400 Wh/mi, I can see the prediction being off.

    I received the firmware update with this feature in the cold of winter and I was WAY below the prediction then. I just chalked it up to the cold and figured it would get better in the spring/summer. It did get better, but still not close to the prediction.
     
  7. Zextraterrestrial

    Joined:
    Mar 11, 2010
    Messages:
    3,636
    Location:
    Humboldt/Los Altos
    maybe your degradation is real :frown:
    don't know what else to make of that
     
  8. apacheguy

    apacheguy Sig 255, VIN 320

    Joined:
    Oct 21, 2012
    Messages:
    3,713
    Location:
    So Cal
    Yeah good point. I wonder if the trip planner estimates capacity or assumes a like-new battery.
     
  9. cantdecide

    cantdecide Member

    Joined:
    Dec 21, 2012
    Messages:
    271
    Location:
    Portola Valley, ca
    You got behind in the first 2 miles, then kept up after that. Tesla sometimes doesn't get things like estimated speed or slope or pavement quite right which could contribute.
    Otherwise I don't know how much Tesla takes into consideration p versus nonp or wheel size, or type of tire.
    I doubt battery degradation would matter as the metering tries to take account of that... It seems to calculate power use mostly separate from battery state.
     
  10. Zextraterrestrial

    Joined:
    Mar 11, 2010
    Messages:
    3,636
    Location:
    Humboldt/Los Altos
    P/ not P makes no difference. Tires will get you better Whr/km (or mi) but that is shown in the usage. and the low power used in the trip meter should have resulted in the green line very much above the grey line
    If my S was behind in the first 2 miles like that and running at 255Whr/mi, I would not be behind, but well ahead very shortly after. The trip prediction is closer to ideal than rated from what I have seen.
     
  11. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,565
    Location:
    Toronto, ON
    I actually wondered that myself. Despite every balancing and "range restoration" trick in the book, the best I currently get is 219 Rated at 90% and 242 Rated at 100%. I got 265 on the nose at 100% when the car was near new. On the other hand, as @cantdecide suggests, I would have thought the algorithm would take all this into account.
     
  12. cantdecide

    cantdecide Member

    Joined:
    Dec 21, 2012
    Messages:
    271
    Location:
    Portola Valley, ca
    Why would p non P not make a difference? Doesn't that make a significant range difference in at least a new model S?
    242 sounds low to me... I'm over 2 years and 38kmiles and still 259. Perhaps there is something with either your battery, battery measurement etc that is sub par.
    Battery measurement is a difficult and complex science and sometimes the Tesla seems to get it wrong... And gradually correct later... Like the time I drove for 10 miles up hill and retained 100% the whole time.
     
  13. MichFin

    MichFin Member

    Joined:
    May 8, 2015
    Messages:
    281
    Location:
    Detroit, MI
    It could be that the maps have inaccurate information about elevation changes in your area. Also, the trip planner is not based on rated range but rather speed limit on the roads your're on. So for example if you're on back roads then it would expect much better than 300w per mile.
     
  14. mknox

    mknox Well-Known Member

    Joined:
    Aug 7, 2012
    Messages:
    8,565
    Location:
    Toronto, ON
    Actually, no. The differences are between the D and the non-D cars. If you drive a P like a non-P you supposedly get the same efficiency.

    I have brought it to the attention of my Service Center and they say the battery is operating normally and as designed.
     
  15. Cyclone

    Cyclone Active Member

    Joined:
    Jan 12, 2015
    Messages:
    4,396
    Location:
    Charlotte, NC
    I sometimes beat it, not always. I always beat it when I try though.
     
  16. Max*

    Max* Autopilot != Autonomous

    Joined:
    Apr 8, 2015
    Messages:
    4,871
    Location:
    NoVa
    Define try?

    If I have range mode on, and do less than or equal to 70mph on a 100+ mile trip, I ALWAYS beat it. If I do 75mph, I SOMETIMES beat it under the same conditions. If I do 80mph, I never beat it.

    [last 3 months of driving. I guess winter may be different, but you've had your car about the same time as me]
     
  17. Cyclone

    Cyclone Active Member

    Joined:
    Jan 12, 2015
    Messages:
    4,396
    Location:
    Charlotte, NC
    I always have range mode off. Try would mean I drive 55 or below with few stop-and-go accelerations vs. 65-70. City driving absolutely won't beat it with all the stop-and-go and hills around me. But even from city to city, if I go the speed limit, I won't beat the prediction.
     
  18. Max*

    Max* Autopilot != Autonomous

    Joined:
    Apr 8, 2015
    Messages:
    4,871
    Location:
    NoVa
    I wonder if that's to do with torque sleep on the dual motor cars vs. your and mnox's RWD cars...
     
  19. Cyclone

    Cyclone Active Member

    Joined:
    Jan 12, 2015
    Messages:
    4,396
    Location:
    Charlotte, NC
    That I wouldn't put past the algorithm. Perhaps some amount of torque sleep is "baked" into the algorithm and our cars will never do that.
     
  20. Zextraterrestrial

    Joined:
    Mar 11, 2010
    Messages:
    3,636
    Location:
    Humboldt/Los Altos
    actually is based on speeds people drive on the road, not the speed limits from what I was told by someone at Fremont - which matches up with what I have seen driving between Humboldt and SF area.

    From my home to the Petaluma SC is 232 miles and I have a 106% trip based on the trip prediction but can just make it while driving the speed limit + 5-7mph
     

Share This Page