Tuning A Better Route Planner with TeslaFi

Well, most of the time range in a EV doesn’t matter, but if you are planning a road trip where chargers are scare or where you are being aggressive. Or you are just a plain nerd, here’s a way to ping pong between Teslafi.com and Abetterrouteplanner.com to figure out what your actually consumption is.

If you want to do this for your driving here’s the procedure

  1. Subscribe to Teslafi and this cloud application will monitor your actual driving usage. So that you can tune it.
  2. Use Abetterrouteplanner.com to actually plan your routes. You will need a login there and you can used the Save Plan feature to create routes. Note the route names are horribly long, so shorten and add notes in the Saved Plans tab. Click on the pencil to edit. Choose Share and copy the big long URL
  3. When you want to use your routes on route, add a calendar entry with the with the URL in the location spot so that you can access it on your phones and also in the Notes because the Tesla in car application will then show and `’i’ information button and clicking on it gets you to the link and it will appear in the car browser.
  4. Now when you are driving you can do real time comparison of what your charge rates are by changing the display performance to show percentages. So your can see how far off it is in prediction. Hopefully it is pessimistic and you will not run out of charge.
  5. And while you are driving look at the Energy chart a bit to see how the in car predictions from the Tesla software are vs. actual.

Tuning the model by looking at actuals: First Leg

A better route planner is using a statistical model so it will more accurate than the Tesla driving application. For long trips, things you need to tune in Show Settings and Show More Settings are, so to match the model, what I do is go through each parameter and match it with actuals.

  1. Speed. This is hard to estimate, but I usually look at Teslafi to get a sense of this. We are normally about 103% of rated, but this will vary. You can estimate this by looking at the time to estimate time in abetterouteplanner and comparing to actual. So find a longish route and just see how long it took you to actually get there on Teslafi.com by looking the total time traveled. Adjust the reference speed in abetterrouteplanner until it matches. I’ve found that depending on traffic, I do 95% of the limit (eg there are times particularly when the limit is 70 mph when I’m lower than the posted sign. These cars particularly the Model 3 burn a lot more due to wind resistance at high speeds, so I normally try to draft trucks which normally are only going 65 or so).
  2. Temperature. This has a big effect when it’s close to freezing, so look at the forecast to see what the weather is like and I try to match it against the average for the day if I’m doing multiple segments.
  3. Rain or snow. This is the biggest variable I’ve found. They use a 10% and 20% penalty, but I’ve found that for the Model P3D it is a bit more.
  4. Your usage at 65mph. This is the biggest variable. You have to know how you perform in your driving habits vs. others. We are normally quite a bit better than the average Model X driver at 350 vs 389 but much closer with the Model 3 at 270-285 vs 295.
  5. Weight. How much are you carrying. I’m not quite sure what excess weight means, but I estimate how many people are with me an luggage. Having an additional 500 pounds makes a difference.
  6. Headwind. I’m not quite sure how to estimate this one, so I just leave it at zero usually.

Adjust for the next leg and repeat

Now I go to the next leg. This is normally after a supercharger stop:

  1. Look at the actual time that you spent there. There is a tuning parameter which is how long does it take you to stop and hook up. It is set to five minutes, but we rarely take that long. Usually 3 minutes is good.
  2. Then repeat the above, so you will get different averages because of temperature changes, different reference speeds and so forth.
  3. This will get you a feel for how driving changes particularly the consumption between mountains and highway. For instance in our Model 3, with a big load an say three people

Draw conclusions given driving conditions

For our own habits, here is what we found:

  1. Non-drafting highway driving. If we are weaving in and out and not paying particular attention to drafting and driving a little aggressively then 293Wh/mile that is the default is a pretty good estimate. This is true even when your reference speed is 95% of the speed limit because wind is such a big factor on the highways. Interestingly drafting behind a Model X or other slippery car doesn’t make much difference, you want to take advantage of those semi’s if they have the same speed limit or the ubiquitous F150s because they really are terrible aerodynamically and block the wind of you very well.
  2. Rain and snow have a big impact on this, so make sure that you have a good sense for what is going on on the trip. Temperature variations of even 5 degrees don’t seem to matter much.
  3. For mountain roads, your reference speed will be much higher. I’m not sure why but on the mountains we drive, we get reference speeeds more like 110-120% of the speed limit. This makes some sense as 5mph over the limit at 45 miles per hour is more like 112% of reference speed but 5mph over 70mph is obviously much less. Because there is less wind, 285 seemed like a better estimate.
  4. If you are drafting while driving. Then at least for us, we get much better mileage. more like 275Wh/mile. And depending on how much traffic there is 100-103% of reference seems about right.
Share

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: