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

Using trip planner on a recent road trip

Discussion in 'Model S: Battery & Charging' started by cpa, May 14, 2015.

  1. cpa

    cpa Member

    Joined:
    May 17, 2014
    Messages:
    939
    Location:
    Central Valley
    I recently returned from a 13-day round trip from California to Milwaukee, Minneapolis and back. I thought this would be the perfect trial for the trip planning feature and the estimated battery reserve feature. I understand the trip planner is "beta," so I set the bar low for results. Here are my actual results and observations which I hope Tesla will give serious consideration to when they enhance and upgrade the trip planner:

    We set off for Kingman, Arizona the first day along SR99, SR58 and I40. The trip planner was pretty good (despite routing us to Harris Ranch--a needless 90-minute detour plus 15-minute charge) at estimating the stops at Mojave, Barstow and Needles. The total drive time to our destination included the time they suggested for charging. The drive was in relatively normal conditions, and the reserve estimate to reach each of the Superchargers was close enough so as not to be of concern. However, the display said that it was "safe to resume trip" with a 5% buffer calculated to our next charging stop. I was leery of such a thin margin, so we always had at least a 12% buffer for shorter legs and 20% for longer legs. The few minutes extra time taken for additional range in my opinion is cheap insurance against unforeseen circumstances. (More on this later.)

    I charged at Kingman to 82% in the evening before retiring. When we started our journey the next morning and set the trip calculator for Durango, Colorado, it directed us to return to the Kingman SC for a 5-minute charge in order to reach Flagstaff even though the trip estimate indicated a 15% reserve upon arrival. This detour feature was becoming annoying, so I canceled the trip planner feature for the remainder of the journey. We took a scenic route through southern Colorado, over Wolf Creek Pass and into Walsenburg before heading northeast on Colorado 10 and then due north to the SC at Limon. I highly recommend this scenic route off the SC highway as it was easy driving with stops in Pagosa Springs (70A) and Alamosa (40A) for two-one hour plug-ins.

    Interstate 70 across eastern Colorado and Kansas was a breeze; EVTripplanner nailed the usage within 4%. We decided to take highway 54 northeast from Columbia, Missouri and reach Springfield SC from the west via I72, and a 4-hour L2 charge was enough range to reach Springfield from Columbia easily.

    My return trip was supposed to be along I90 through South Dakota and then angling northwest through Broadus, Montana and charging at the RV park for several hours before reaching Billings. However, due to snow and temperatures in the 20s in Rapid City (and even as far south as Fort Collins) I decided to return via Interstate 70 through Worthington and Council Bluffs.

    Things got interesting driving west through Kansas the next day. I charged to 90% at Topeka to reach Salina, and made it driving 65 the entire way with 39%, averaging 340 wh/mile. I knew there was elevation gain from Salina to Limon, and it was getting breezy in the late morning, so I charged to 96% at Salina to reach Hays, 94 miles west. My expected buffer was 29%, so I felt confident! Wrong! Thirty miles into the trip the reserve had dropped to 18%, so I slowed down from 65 to 58, and I reached Hays with 5%. I then max charged at Hays to reach Goodland, 143 miles west. The estimator indicated a 32% reserve when I left. By now the wind was 30+ MPH from the west, so I figured I would start out at 60 MPH driving into the headwind. Again, the reserve dropped precipitously after 20 miles, so I kept lowering my speed down to 48 MPH for the last 80 miles. The reserve finally leveled out at 9% reserve with 60 miles to go, and I reached Goodland with 7%. These two legs took almost 420 wh/mile! I thought that I had factored enough buffer to account for both elevation gain and headwind, but I missed badly.

    I have concluded from my lone experience that the trip planning feature needs to expand its calculations by having some sort of input from the driver to account for spontaneous changes in weather or temperature. I presume the software knows the 1,500-foot elevation gain over the 143-mile drive from Hays to Goodland. The software needs to have the driver input options like estimated wind speed and ambient temperature in order to consider those factors on range. Once the calculation is determined, the driver should be notified as to the estimated wh/mile at different speeds--so the driver can make a reasonable determination to keep his speed early in the drive at 50 or 55 to have a reasonable buffer. The software might be able to notify us that our speed is good or whether to slow down from 55 to 50 or speed up from 55 to 58 as circumstances dictate. I can imagine new owners being totally naive about this complicated issue and wind up stranded and very upset that they ran out of juice 35 miles away from any charging spot, let alone their intended Supercharger.

    Needless to say, my crawling along Interstate 70 at 48 MPH for 60 miles was not a good public relations campaign for driving BEVs!

    One final observation about the navigation program: My last leg to return home was to get from Reno to the Manteca SC by taking US395 south to Minden, then taking SR88 over Carson Pass to Stockton and then onto Manteca. The program first routed me back to Truckee and then down Interstate 80. I tried again south of Carson City, and it routed me to US50 to South Shore and into Sacramento. At Minden, it routed me over Kingsbury Grade to South Shore and Sacramento, again on US50. At Woodfords, it routed me north on 89 to Meyers and US50 and into Sacramento. It wasn't until I had surmounted Carson Pass that it decided I really knew where I was going and kept me on SR88. We users should be able to select our route(s) among different choices presented instead of one pre-determined route that focuses solely on Superchargers.

    So, my suggestion to Tesla is that until Superchargers or convenient L3 charging/80A L2 opportunities are ubiquitous throughout the country where we are never more than 60 miles away, is that we need some sort of user input to guide us into how much to charge and at what speed to drive until we are in the clear.
     
  2. MorrisonHiker

    MorrisonHiker Beta Tester

    Joined:
    Mar 8, 2015
    Messages:
    1,911
    Location:
    Morrison, Colorado, USA
    Thanks for the details concerning your recent trip. Maybe I missed it, but do you have a 60kWh or 85kWh battery? (I'm in Tapatalk so I'm not seeing what's in your signature.)

    I often make the drive across Colorado and Kansas and hope to do it in a Tesla soon...but I would hate to have to do any of it at 48 mph!
     
  3. Blu Zap

    Blu Zap Grinning member

    Joined:
    Oct 1, 2014
    Messages:
    437
    Location:
    San Rafael, CA
    @CPA
    thanks for this very informative trip summary. It helps us to figure out factors that effect long range trips. I am surprised on the return trip weather and wind factors were not factored in on the range calculation. I had heard that they were, so understanding, it is not the case. Also the routing back home over the Sierras, at some point it should allow you to pick your alternative route, like Google Maps does. Both obvious improvements needed. Again, thanks for your post.
     
  4. napabill

    napabill Active Member

    Joined:
    Apr 30, 2009
    Messages:
    1,325
    Location:
    Napa, California, United States
    The TripPlanner software works pretty well UNLESS there are multiple SC's along the route. I have found that the only solution currently is to de-activate TripPlanner and just select the SC you want to go to. That works fine.
     
  5. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    See my description at How to read the Energy chart? on how to use the Energy/Trip screen. I think that is the only useful part at this point. As the OP said, be very careful of headwinds and cold.
     
  6. FredTMC

    FredTMC Model S VIN #4925

    Joined:
    Dec 26, 2012
    Messages:
    3,003
    Location:
    Orange County CA
    I'm very confused by the OPs post because I thought Trip Planner takes into account Wind Speed (in addition to Elevation,Temp and speed). If Trip Planner can't predict accurately, it's useless

    i don't have tech package. I've tried in on a loaner and it worked okay. But never put it to the test on a long trip
     
  7. Johann Koeber

    Johann Koeber Member

    Joined:
    May 1, 2012
    Messages:
    663
    Location:
    Nuremburg, Germany
    Thanks for the detailed report.

    The software should need no user input. All the data is there or easily accessible on the internet.

    Wind speed?
    average consumption?

    The only user input I would like to see is the amount of buffer - this might be a personal preference. And I share your desire for 20 %.

    I'm glad you made it. In 84.000 km with the moden S I never got stuck, but I am more adventurous now, than I was initially.
     
  8. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    My experience is that the energy prediction of the trip planner does not take current weather conditions into the estimate while you are charging. Once you are on the road, the Energy:Trip page does seem to do a pretty good job of comparing your energy usage over the last 5-10 miles with it's model and project a SoC at the destination.

    In the trip planner's present incarnation, the user has to add his own buffer to the trip planners estimate while charging. Once driving a few miles, the projected estimate at the destination is pretty good. Be careful how much buffer to give yourself before leaving the charger!

    BTW, my favorite web page for wind visualization that runs on smart phones, but not the Tesla web client is Windyty!
     
  9. gregincal

    gregincal Active Member

    Joined:
    Oct 26, 2012
    Messages:
    2,137
    Location:
    Santa Cruz, CA
    It's easy to underestimate the effects of headwind. Driving into a 30 mph headwind at freeway speeds doesn't just require a little more energy, it can easily require 60% more energy (just look at the charts for wh/mile and compare 70mph with 100mph).
     
  10. FredTMC

    FredTMC Model S VIN #4925

    Joined:
    Dec 26, 2012
    Messages:
    3,003
    Location:
    Orange County CA
    yep. That's my point! Why isn't wind correctly modeled in trip planner. Wind DOES make a huge difference
     
  11. Khatsalano

    Khatsalano Member

    Joined:
    Mar 21, 2015
    Messages:
    648
    Location:
    San Mateo, CA
    Thank you for the trip report. I have learned to distrust 6.2 Nav and rely on Waze + EVtripplanner.com. Glad you made it ok, but I'd be pretty scared at 48mph on the interstate!

    - K
     
  12. CHG-ON

    CHG-ON Still in love after all these miles

    Joined:
    Jun 24, 2014
    Messages:
    2,766
    Location:
    Santa Cruz Mountains, USA
    Great report! I am planning a trip from SF bay area to Park City, UT and have been modeling my energy usage, targeting no less than 25% buffer because I don't trust the trip planner yet. I've been using EVtripplanner and creating my own charging targets based on that data. It'll be interesting to compare my estimates against EV, Tesla's trip planner and the actuals. If I can do the trip, I'll post a detailed comparison of everything.
     
  13. Panoz

    Panoz Member

    Joined:
    Nov 19, 2013
    Messages:
    303
    Location:
    Colorado Springs, CO
    This is a very educational thread, I had no idea the reserve would change that much based upon local weather conditions. And yes, 48 mph on the Interstate is scary.
     
  14. Duma

    Duma Member

    Joined:
    Nov 16, 2014
    Messages:
    150
    Location:
    Pittsburgh, PA
    Based on a recent trip with 3 superchargers along the route, the TripPlanner software did OK but the algorithm definitely still needs some work. In order to minimize charging time at each SC, the algorithm tries to keep the SoC as low as possible. This resulted in including one nuisance stop to charge for 5 minutes that was avoidable by charging a bit longer at the preceding stop. Seems like the algorithm fails to fully account for the extra driving time needed to route off the highway to the SC (in this case the SC was along the optimal route so there no additional impact of a suboptimal route.)

    There also seem to be some bugs. For reasons that are not important, ended up restarting the (remainder of the) route while near a SC where I had just charged to 80% SoC. TripPlanner wanted to take me back to that SC for a 0 minute charge and then routed me to the wrong SC on the path to my destination, which is so wrong that I have to conclude there is bug. (Note that the original plan from the start of the trip had selected the correct two SC, after bypassing an unnecessary SC because I was starting from home with enough SoC.)

    The frustrating part is that it is not possible to remove an unnecessary SC stop. As napabill said, the only work around is to manually select individual SC one step at a time.

    The trip planning software is clearly using waypoints internally to manage the intermediate SC charging stops. What remains unclear is why the current algorithm is so inflexible that we as users are not permitted to delete, add, or rearrange waypoints. Obviously manual changes could result in an "invalid" (too much distance between charging points) route, but the solution would be to warn and/or recalculate the needed charging stops, not to completely forbid changes. Hopefully this will all get resolved over time and is not a permanent limitation due to constraints in the interface between the Navigon and Google Maps parts of the solution.
     
  15. Owner

    Owner Active Member

    Joined:
    Dec 20, 2012
    Messages:
    1,229
    Location:
    San Francisco Bay Area
  16. cpa

    cpa Member

    Joined:
    May 17, 2014
    Messages:
    939
    Location:
    Central Valley
    I charged at the HPWC at the Peppermill for about 2 hours. I left with 225 miles of rated range to take me over Carson Pass and Spur on SR88 before reaching Manteca with 60 miles in the battery. The strong headwinds blowing north as I drove south on US395 kept my speed at about 55 to be extra safe. Since the navigation feature kept wanting to route me to US50, I was mostly in the dark as to how fast I could climb Carson Pass once I reached Woodfords and not run out of juice. I used 46kWh of electricity for that leg, averaging 233 wh/mile. From the summit to Stockton, I actually gained 1% of battery capacity, as that stretch is all downhill.

    Your report is quite accurate, and I agree with all your observations. I too wonder about the efficacy of Tesla's logic when more and more people buy Teslas and drive them like ICE vehicles because they are ignorant and naive about the nuances of driving a pure BEV to destinations that are either beyond charging locations or the weather is so severe that energy usage is 1/3 greater than under normal conditions.
     
  17. Owner

    Owner Active Member

    Joined:
    Dec 20, 2012
    Messages:
    1,229
    Location:
    San Francisco Bay Area
  18. jerry33

    jerry33 S85 - VIN:P05130 - 3/2/13

    Joined:
    Mar 8, 2012
    Messages:
    12,753
    Location:
    Texas
    When I used it over the weekend it wanted to give me zero % left when I arrived that the next one. In some cases it said I had enough energy after five minutes of charging. I just waited until it showed 30%. Running it to zero doesn't appear to make a lot of sense--especially because there was some major construction congestion.
     
  19. cpa

    cpa Member

    Joined:
    May 17, 2014
    Messages:
    939
    Location:
    Central Valley
    Yes, I remember vividly. The first episode if I recall occurred in northwest Nevada where California 299 peters out east of Cedarville at the Nevada line and turns into a dirt road (it used to be Nevada Highway 3A back in the day, but may have had its signage removed.) The second one was in Southern Oregon. (And I did not peek at your links either!)
     
  20. artsci

    artsci Sponsor

    Joined:
    May 10, 2012
    Messages:
    4,675
    Location:
    Timonium, Maryland
    My experience with the beta on our coast to coast drive to TMC Connect has been a total disaster. First, I did not know the thing was turned on, which led to abject confusion about why the nav system was behaving so oddly.

    To start, we chose the Hagerstown supercharger as our first stop. It's due west -- the direction we needed to go. The trip planner attempted to divert us to Bethesda, which is southwest in the wrong direction. No matter what we did we could not get the nav system to stop trying to guide us in that direction. This happened again and again -- the trip planner taking us to the wrong supercharger and refusing to be overridden. Finally, when we got to Illinois we figured out what was going on and just turned the damned thing off. No problem since.

    My advice would be to shut the thing off until Tesla irons out the obvious flaws. For now, based on my experience, it's useless. Not only that it will attempt to divert you from a carefully planned route, which ours is.

    I think Tesla should just adopt EV Trip Planner. It's far better than what Tesla is doing on its own.
     

Share This Page