EV Trip Planner or ??

[Ulmo]

OT, but I want to point out how there's a lot of obvious ways Tesla could improve their product line for relatively small amounts of money. This is one example: EV Trip Planner has been an obviously better information source than Tesla's on-board systems for many years running now. Tesla could purchase the quality, database and logic behind EV Trip Planner, and/or help them and integrate with it on-board. Alternatively, Tesla could in-house the same quality level. I feel like Tesla has been hiring people who do not have the same top-notch capabilities that a lot of their customers and enthusiasts have, or their corporate culture is getting in the way. I realize that Tesla has a lot on their plate and is having some financial priorities to get in line since they are a hugely growing company, but in my opinion, fixing these things will end up costing Tesla less down the road in customer support, happiness, margins, etc.. I'm fully aware that Tesla might have some conflicts of interest that EV Trip Planner might not have (such as the backend databases, etc.), and that Tesla has hugely improved their offering for this data. But, it also continues to be an "opportunity for improvement". I'm also aware that in a few years, there will be so many SuperCharger's that the use parameters for these options will have changed.

I'm very happy that both options are available for those who are in Tesla's. I wish I had a Tesla to enjoy myself with these tools. I commonly drive long distance, and have been wondering what it would be like to use such systems. I've loved playing with EV Trip Planner to plan out many of my common trips. I've often learned of possibilities I wouldn't have considered myself. (Lately, though, I've consigned myself to the notion I won't need it for many years since I can't afford one and by then the landscape will have changed so much that I might as well stop planning and wait; when there's SuperChargers every 50 miles, rules of thumb and winging it will become much easier.)

 

[ZBB]

I could tweak the numbers I use but my understanding was that speed 1 was driving the speed limit, and that's what I do (no point going faster than the speed limit where I drive plus my best trip speed in an S60 is probably 60-65 mph, so going faster means a longer trip due to longer Supercharging times).

1 = the average speed as reported through Google's API. I use 1.1 as my setting -- basically builds in a hedge factor. I'm also in a S60 by the way, and usually drive about 5 over (unless I really need the range, then I drop back to the limit). The Supercharger charges faster than I drive (most sessions average 150 mph+ for a charge up to 90-95%), so I'll spend the extra time charging...

This is key: the Tesla trip energy gauge is really quite good now, unlike when you were a pioneer in doing long trips. Now it is simple to use nav to select the next Supercharger station, charge until the energy plot shows enough to make the trip, plus whatever buffer one is comfortable with, and then just go when the charge hits that level. It really is a huge improvement over the early days, from what I can tell. And the business of nav routing back to a Supercharger if one is cutting it close can now be turned off. So, I am really pleased with the current navigation to Supercharger Stations and the trip energy plot. It is very easy to use and works well IME.

My last trip had a couple legs where the car was bouncing between estimates -- would show me arriving at a Supercharger with 5% for a couple min, then drop down to right around 0% (with the warnings to slow down to reach destination…). I was driving the speed limit with cruise control on. That's still too much variation for me...

 

[Tree95]

My last trip had a couple legs where the car was bouncing between estimates -- would show me arriving at a Supercharger with 5% for a couple min, then drop down to right around 0% (with the warnings to slow down to reach destination…). I was driving the speed limit with cruise control on. That's still too much variation for me...

Yep, my last trip about a month ago showed the same thing - abrupt changes by more than 5% even though the car was keeping pretty much a steady speed. I've drivin 34K miles in a year with probably 20K of that between superchargers, and this behavior was new for me. It's been wrong before, it's told me to turn around the moment I leave a supercharger, it's had all sorts of issues (most fixed), but this was the first time I saw the estimated % charge estimation jump by large increments up and down every few minutes. It sort of undermines any trust I had.

 

[green1]

I set the speed to 1.2, and then charge to that number. Builds in the 20% charge buffer that I'd otherwise build in manually.

 

[dgpcolorado]

1 = the average speed as reported through Google's API. I use 1.1 as my setting -- basically builds in a hedge factor. I'm also in a S60 by the way, and usually drive about 5 over (unless I really need the range, then I drop back to the limit). The Supercharger charges faster than I drive (most sessions average 150 mph+ for a charge up to 90-95%), so I'll spend the extra time charging...

The mph charging speed is not the same thing as optimum travel time. The reason is that increasing speed takes a lot more energy due to drag, which increases at the square of velocity. And charging speed is (a lot) faster when the battery is nearly empty than it is when half full or higher as the charge rate tapers drastically.

Add those two things together and there is an optimum speed for total travel time (driving plus charging) and it is not related to average charging speed because neither energy use while driving at highway speeds nor charging speed are linear functions. For an 85 I've seen it said that the optimum speed under mild weather conditions is in the 70-75 mph range. For a much slower charging 60 it figures to be a lot lower than that, although I haven't tried to do the calculation. The upshot is that driving faster than that optimum speed leads to slower overall travel time. As has been discussed by others in other threads, the best strategy is to arrive at the next Supercharger Station with a near empty battery and charge just enough, plus a small buffer, to make the next one. And drive at the optimum speed, whatever is is for the car and conditions. Unless one is going to take a long break for a meal stop or something, this charging regimen will give the shortest travel time.

The speed limit on large portions of my upcoming trip is 80 mph. If I drive that speed I will spend longer in extra charging, from the much higher energy use of high speed driving, than the small amount of time I save by traveling that fast. I'll probably compromise by driving at 70-75, which will also mean longer charging times, but not by so much. I have found that I do need the time to walk around during charging breaks.

I met a new Model X driver at the Denver Supercharger Station recently. He had driven mostly at 85 mph across the Great Plains and was unaware of the concept of optimum speed and charging time. His family was getting restless at the long charge times he was using (and he had already charged more than double what he needed for his final trip leg, due to inexperience). He could have slowed down, charged for much shorter times, and decreased his total travel time significantly. (Although I can understand the temptation to cross the Plains as fast as possible; it is very boring driving.) But he was new to Tesla and still learning; this was his first road trip. I expect that he will improve over time.

 

[1101011]

OT, but I want to point out how there's a lot of obvious ways Tesla could improve their product line for relatively small amounts of money. This is one example: EV Trip Planner has been an obviously better information source than Tesla's on-board systems for many years running now. Tesla could purchase the quality, database and logic behind EV Trip Planner, and/or help them and integrate with it on-board.

It would seem to make sense for Tesla to license EVTripplanner.

 

[CygnusX1]

When I use EVTP for round trips like from A to B to A, it doesn't show SuC stops even if the total amount of kWhrs is 80 or 100... And I have set up a S 70.

Why is that? Am I doing something wrong? Anyone else encountered that behavior?

Cheers

 

[Boatguy]

I just tried a round trip and it seemed to put the SC stops in the right place, but maybe there is some issue with round trips. Here are a couple of suggestions:

If you've tried several routes, try refreshing the web page before trying a new route.

Make sure you hit the "Route through Superchargers" button and not "Route Direct".

Instead of doing a pure round trip, try making your final destination a city very close to the starting city.

 

[ZBB]

The mph charging speed is not the same thing as optimum travel time. The reason is that increasing speed takes a lot more energy due to drag, which increases at the square of velocity. And charging speed is (a lot) faster when the battery is nearly empty than it is when half full or higher as the charge rate tapers drastically.

Add those two things together and there is an optimum speed for total travel time (driving plus charging) and it is not related to average charging speed because neither energy use while driving at highway speeds nor charging speed are linear functions. For an 85 I've seen it said that the optimum speed under mild weather conditions is in the 70-75 mph range. For a much slower charging 60 it figures to be a lot lower than that, although I haven't tried to do the calculation. The upshot is that driving faster than that optimum speed leads to slower overall travel time. As has been discussed by others in other threads, the best strategy is to arrive at the next Supercharger Station with a near empty battery and charge just enough, plus a small buffer, to make the next one. And drive at the optimum speed, whatever is is for the car and conditions. Unless one is going to take a long break for a meal stop or something, this charging regimen will give the shortest travel time.

The speed limit on large portions of my upcoming trip is 80 mph. If I drive that speed I will spend longer in extra charging, from the much higher energy use of high speed driving, than the small amount of time I save by traveling that fast. I'll probably compromise by driving at 70-75, which will also mean longer charging times, but not by so much. I have found that I do need the time to walk around during charging breaks.

I've posted my Supercharging time table for an S60 a couple times now. Might as well post again for reference -- you might find it handy. I built this off data logged during my first couple Supercharger visits, and its proven fairly accurate since (I've supercharged 66 times now…).
View attachment 85740

As for optimal time… I don't disagree with the concept of optimizing total charging and driving time, but I do disagree that driving slower helps optimize drive time when supercharging. Here's an example -- using from the Buckeye to Quartzite AZ Superchargers, which is 100 actual miles and speed limit is mostly 75 on that part of I-10. Using EV Trip Planner and my charging estimate table, I get:

• At 1.1 speed factor, EVTP estimates using 132 rated miles with an average speed of 79 mph. Drive time is 1 hour 15 minutes
• At 1.0 speed factor, the estimate is 117 RM with average speed of 72 mph and a 1 hour 23 min drive time
• Assuming I left Buckeye at 190 RM -- I'd arrive with 58RM driving faster. So charging back to 190 (enough to get to Indio with buffer would take ~56 minutes. Drive and charge time would be 2 hours 10 minutes.
• Driving slower, I would arrive with 73RM, and it would take ~52 minutes to charge back to 190. Drive and charge time would be 2 hours 15 minutes
• So -- driving faster here is still net quicker… 5 minutes!

 

[Tree95]

Yep, my last trip about a month ago showed the same thing - abrupt changes by more than 5% even though the car was keeping pretty much a steady speed.

Wow. Drove today thru Tahoe, down 395 Topaz Lake and the nearly complete Hawthorne SC.

The battery range/charge graph jumps all over the place. Hills are horrible. The energy use goes up to climb, and the nav system assumes this high energy consumption will continue for the next 100 miles, and it has an utter shitfit going from 20% estimated at the destination to telling me to slow to below 55. When the car levels off, a mile or two later, it goes back to 20%. Once, after a grade, it remained at below zero. So, i canceled, and immediately reentered the same destination, and it's back to estimating 18%. Jesus, this is a massive step backwards from previous behaviour.

Something I noticed about 2 months ago on a trip through the Mojave is that the nav system way underestimates the power needed for a long uphill grade. A year ago, it estimated grades just fine. Now, after a long climb from the Central Valley up to the high desert, the green actual line vs grey predicted line are off by 15-20%. Yes, I'm doing the speed limit; speeding up hills is a killer for range. It was 50-55, dry roads.

While the egregious bugs like telling you to turn around immediately after leaving an SC are gone (and were easily worked around by touching the 'remove all superchargers') the actual range/charge prediction that used to work reasonably well for me has become pretty untrustworthy.

 

[dgpcolorado]

I've posted my Supercharging time table for an S60 a couple times now. Might as well post again for reference -- you might find it handy. I built this off data logged during my first couple Supercharger visits, and its proven fairly accurate since (I've supercharged 66 times now…).
View attachment 85740

Interesting chart, must have taken a lot of work to assemble! My car was a lot slower at the Moab Supercharger; it took me more than 90 minutes to get from about 30 RM to 90%. Whether that's because Moab is slow or because my battery might not have been balanced, I couldn't say.

As for optimal time… I don't disagree with the concept of optimizing total charging and driving time, but I do disagree that driving slower helps optimize drive time when supercharging. Here's an example -- using from the Buckeye to Quartzite AZ Superchargers, which is 100 actual miles and speed limit is mostly 75 on that part of I-10. Using EV Trip Planner and my charging estimate table, I get:

• At 1.1 speed factor, EVTP estimates using 132 rated miles with an average speed of 79 mph. Drive time is 1 hour 15 minutes
• At 1.0 speed factor, the estimate is 117 RM with average speed of 72 mph and a 1 hour 23 min drive time
• Assuming I left Buckeye at 190 RM -- I'd arrive with 58RM driving faster. So charging back to 190 (enough to get to Indio with buffer would take ~56 minutes. Drive and charge time would be 2 hours 10 minutes.
• Driving slower, I would arrive with 73RM, and it would take ~52 minutes to charge back to 190. Drive and charge time would be 2 hours 15 minutes
• So -- driving faster here is still net quicker… 5 minutes!

Interesting, point taken (and appreciated). Using EVtripplanner numbers but cutting the buffers down I also get numbers that are pretty much a wash for speed 1 and speed 1.1. (Although since my charging speeds seem slower than yours I need to measure them some more, since that would make a difference.)

—> At 0.9 speed I get 102 RM, 1:36 drive time, for the first leg from Buckeye SpC to Quartzite SpC, using default temp and payload, and 114 RM, 1:58 drive time, for the Quartzite to Indio SpC leg.

So, arrive at Buckeye at 20 miles, charge to 122 RM, including a buffer of 20 RM, would take about 25 minutes from your chart. Arrive at Quartzite at 20 RM and charging to 134 RM, including a buffer of 20 RM, would take about 27 minutes for the Indio leg.
Buckeye to Quartzite would be the charge time of 25 minutes plus drive time of 1 hour 36 minutes = 2 hr 1 minute
Quartzite to Indio would be the charge time of 27 minutes plus drive time of 1 hour 58 minutes = 2 hr 25 minutes

—> At 1.0 speed I get 114 RM, 1:27 drive time, for the first leg from Buckeye to Quartzite, and 128 RM, 1:46 drive time, for the Quartzite to Indio leg.

So, arrive at Buckeye at 20 miles, charge to 134 RM for Speed 1, would take about 29 minutes from your chart. Arrive at Quartzite at 20 RM and charging to 148 RM, including a buffer of 20 RM, would take about 35 minutes for the Indio leg.
Buckeye to Quartzite would be the charge time of 29 minutes plus drive time of 1 hour 27 minutes = 1 hr 56 minutes
Quartzite to Indio would be the charge time of 35 minutes plus drive time of 1 hour 46 minutes = 2 hr 21 minutes

—> At 1.1 speed I get 129 RM, 1:19 drive time, for the first leg and 144 RM, 1:37 drive time, for the second leg.

Arrive at Buckeye at 20 miles, charge to 149 RM, including the buffer, in 36 minutes. Arrive at Quartzite at 20 RM and charge to 164 RM, including the buffer, in about 45 minutes to make the leg to Indio.

Buckeye to Quartzite would be the charge time of 36 minutes plus drive time of 1 hour 19 minutes = 1 hr 55 minutes
Quartzite to Indio would be the charge time of 45 minutes plus drive time of 1 hour 37 minutes = 2 hr 22 minutes

—> At 1.2 speed I get 145 RM, 1:12 drive time, for the first leg and 161 RM, 1:29 drive time, for the second leg.

Arrive at Buckeye at 20 miles, charge to 165 RM, including the buffer, in 45 minutes. Arrive at Quartzite at 20 RM and charge to 181 RM, including the buffer, in about 55 minutes to make the leg to Indio.

Buckeye to Quartzite would be the charge time of 45 minutes plus drive time of 1 hour 12 minutes = 1 hr 57 minutes
Quartzite to Indio would be the charge time of 55 minutes plus drive time of 1 hour 29 minutes = 2 hr 24 minutes

—> At 1.3 speed I get 162 RM, 1:07 drive time, for the first leg and 180 RM, 1:22 drive time, for the second leg.

Arrive at Buckeye at 20 miles, charge to 182 RM, including the buffer, in 56 minutes. Arrive at Quartzite at 20 RM and charge to 200 RM, including the buffer, in about 83 minutes to make the leg to Indio.

Buckeye to Quartzite would be the charge time of 56 minutes plus drive time of 1 hour 7 minutes = 2 hr 3 minutes
Quartzite to Indio would be the charge time of 83 minutes plus drive time of 1 hour 22 minutes = 2 hr 45 minutes

 

[Steve2498]

Tre95:

What's worse is I've found it often (usually) very good at predicting, but not always and I'm not always sure when to trust it and when not to. The result is on new routes I plan on higher reserves.

On a recent trip from the Bay Area to Seattle and back, the entire trip under ideal conditions, for all but one leg the planner was within 2% of actual usage. If I could count on that I'd use 10% or 15% reserve all the time. But on the segment (which included a lot of elevation changes) where it failed me, it was off by 20%! I'm glad I used 30% reserve on that leg.

 

[dgpcolorado]

Wow. Drove today thru Tahoe, down 395 Topaz Lake and the nearly complete Hawthorne SC.

The battery range/charge graph jumps all over the place. Hills are horrible. The energy use goes up to climb, and the nav system assumes this high energy consumption will continue for the next 100 miles, and it has an utter shitfit going from 20% estimated at the destination to telling me to slow to below 55. When the car levels off, a mile or two later, it goes back to 20%. Once, after a grade, it remained at below zero. So, i canceled, and immediately reentered the same destination, and it's back to estimating 18%. Jesus, this is a massive step backwards from previous behaviour.

Something I noticed about 2 months ago on a trip through the Mojave is that the nav system way underestimates the power needed for a long uphill grade. A year ago, it estimated grades just fine. Now, after a long climb from the Central Valley up to the high desert, the green actual line vs grey predicted line are off by 15-20%. Yes, I'm doing the speed limit; speeding up hills is a killer for range. It was 50-55, dry roads.

While the egregious bugs like telling you to turn around immediately after leaving an SC are gone (and were easily worked around by touching the 'remove all superchargers') the actual range/charge prediction that used to work reasonably well for me has become pretty untrustworthy.

Have you had one of the recent firmware updates? The reason I ask is that my car energy projection works fine on steep hills and mountain passes (when I drive the speed limit) and if your issues are the result of an update I'll skip it! I'm currently on 2.16.17 and it seems to be working fine, so I guess that I shouldn't mess with it.

 

[Tree95]

Have you had one of the recent firmware updates? T

2.16.37

 

[Petra]

I tend to sanity check my trip plans at home using EV Trip Planner before I leave and use Tesla's system while on the road, keeping in mind what I had worked out with EVTP. The sanity check is important for me because Tesla's system is still pretty rough... it generally works well, but will occasionally decide that I can't make it to my destination and try to reroute to the nearest Supercharger when I know full well that my destination is within range. Some of the stranger bugs don't help, like when it suddenly drops the estimated arrival SoC from something reasonable to a negative value, freaks out, crashes, and reboots the MCU. While that particular fault has only happened once, it was very much an 'ignore Tesla trip planner and mute Betty for the rest of the trip' moment.

 

[HankLloydRight]

Just to plug the site I built for "after the fact" Supercharge trip analysis -- SuperCharger Trip Logger

The intention was to build up a large database of actual SC->SC trip logs, so anyone could lookup what other people experienced under "real life" conditions (i.e. not estimated like EVTripPlanner), including traffic, weather, temperatures, etc between any selected Superchargers. But it never really caught on, so I don't have a lot of data in it yet.

I have logged dozens of my own trips between NY and Florida, North Carolina, Chicago, Ohio, and Boston, so I can now go back and see what my actual speed and consumption was for each leg, so when I go back to repeat a trip, I bring up the previous trips to see what I did to help me plan.

All that it requires is that you log your remaining rated miles, distance traveled, and energy used at the start and end of each SC stop using the touchscreen or any mobile or table browser. Once those three items are input, it calculates the rest.

Here is an example of the output of a two-day trip between Florida and CT:

 

[dgpcolorado]

2.16.37

Thanks. I think I'll skip any updates for awhile, at least until feedback suggests that they didn't break the trip energy plot. Just in case, since I have a very long trip planned.