Jump to content

Bo (ABRP)

Administrators
  • Content Count

    154
  • Joined

  • Last visited

  • Days Won

    15

Everything posted by Bo (ABRP)

  1. The Tesla API is a bit shaky and from time to time it will just return a 408 request timeout for no obvious reason. I just tried logging on to MyTesla on both Chrome PC and Chrome Android while being logged in to ABRP, and it works for me. In the first case it took like 10 seconds for the dot to turn green, though.
  2. Bo (ABRP)

    Waypoints Not Working

    Actually, that last example is not a bug in itself. When you add a charger which is already in the plan as a waypoint and click the waypoint options, it will propose to hard-set the charging time and power to what was in the plan. If you leave these numbers as they are, ABRP will obey these numbers and not charge longer (or shorter). Therefore, when you add 60% arrival, it will have to go via Albany to pump up the SoC to meet your arrival SoC 🙂 Here is the plan without forced charging time on Queensbury: https://abetterrouteplanner.com/?plan_uuid=d6263052-b1a6-4144-9787-9d65727a5083
  3. Thanks for feedback! The bug in the Tesla browser is that they do not display the "Allow access to location" dialog - they will instead just report back that the user said "No". As soon as we get that "No" we will look for IP location instead. But I realize that this may be annoying if you are actually logged in with MyTesla. I'll implement that delay you proposed! Edit: Done, see how it works for you.
  4. At least in Tesla SW 2019.16.* there was a bug in the browser which made GPS position fail for web sites. To be able to do anything useful at all, we added IP address positioning, which is unfortunately only useful when you are on wifi. The only solution right now is to log in using your MyTesla account, which is however nice from many other aspects.
  5. Bo (ABRP)

    Waypoints Not Working

    The original plan link above is from "My position" which will then create a different route for everyone clicking on it 🙂 This can be circumvented using and additional &mypos=1 in the URL; https://abetterrouteplanner.com/?plan_uuid=703e3b39-91cf-48f8-87e5-4daab71051a6&mypos=1 (which will use the position you used for "My position"). With that URL the plan works fine for me (as far as I can see). Are you still having issues with it?
  6. I can replicate it using your steps above, but it seems to be very sensitive. If I force it charge 1 minute longer at Kaufland, it does not insert that extra charger, it does not seem to generally happen. I'll keep this particular case in my "things to investigate" folder and if you find any more obvious cases let me know.
  7. Thanks for the report and good debug info. This should now have been solved.
  8. We have a beta.abetterrouteplanner.com for new feature testing, and we will use it more. In this particular case, it was a format change (for e.g. settings) which was one-way so testing and then going back would not work. Sorry for the trouble it caused!
  9. Thanks for the report, I think I fixed this just after you reported it. Sorry for messing thins up, we had to do an interface restructuring to prepare for some pretty cool new features.
  10. This is actually an ABRP feature. The mountain road speeds are not correctly estimated by OSRM (the open source routing backend we use) since it does not take road curvature into account. Google maps says 32 minutes for that stretch, ABRP says 39 minutes. In practice, for this stretch of road, do you think you could drive much faster?
  11. (Edits: Correct charging times for 10-50 kWh for new charge curves, which previously were too optimistic. Added charging time for BTX6 100 kWh battery.) About a month ago, Tesla started rolling out updates to their fleet to allow for faster charging at the V2 Superchargers. These are the classic Superchargers which are everywhere, and they are still physically 145 kW maximum, serving a pair of stalls. The new V3 Superchargers will be able to serve 250 kW, per stall, but they are not yet available - we will report on their performance once we have data. But how does the fast V2 Supercharging actually perform in reality? Thanks to ABRP users logging in with MyTesla in ABRP, and generously sharing data with us, we are able to record charging sessions for a lot of different cars at a lot of different supercharger locations. So this is actual crowdsourced data from the real world. Increasing the charging power on vehicles should of course be a win for the drivers, leading to shorter charging times, but in particularly this case, there are some if's-and-but's: For Model S and X, this is done by just decreasing the margins; will the cars and batteries actually cope with it? The V2 Superchargers can provide 145 kW max for a pair of stalls. Normally, the first car to charge in a pair would get all power it could take, and the rest would be given to the other stall - now how would this work if the first car can actually take 145 kW? Obviously, actually getting full power would be even more dependent on Supercharger occupancy. Since the Supercharger charging cables and connectors were not build to deliver 145 kW, this will make the charging power even more dependent on external factors such as temperature. For these reasons, we have chosen to be cautious and not yet enable the faster charging curves in ABRP. We will do it as soon as we see that it works well enough in the real world. Enough talking, let's get to actual data! Model 3 Long Range, the BT37 battery (75 kWh) The plot below shows charging data gathered from Model 3s in ABRP from April 1st 2019, each sample is a faint blue dot. The red curve is the present (old) charge curve in ABRP for this battery, and as we can see it is limited to about 120 kW. In charging sessions with cars with upgraded charging power, which depends on software version, and at Superchargers which support it, we see a clear cloud of charging points up around 145 kW for 10-40% SoC. The yellow curve is the fitted charging curve to this new data. Translating these charge curves into charging times from 10 kWh to 50 kWh in the battery (in absolute energy to make it comparable between battery sizes) we get: Old 120 kW charging curve: 24 minutes 27 seconds New charging curve: 21 minutes 31 and seconds That is a pretty decent time saving! As long as there is Supercharger power available for you, of course. Clearly, the BT37 battery and Model 3 has been designed from the beginning for higher power charging. Model S/X100, the BTX6 battery (100 kWh) The BTX6 battery is the old grandmaster among Tesla batteries. Except for being the clearly highest capacity battery, it could also take 120 kW longer than any other battery. However, in the plot below, we notice something when it is pushed to 145 kW - see those decreasing lines of charging dots from 145 to 120 kW? It really seems like the charging speed tapers off pretty quickly irrespective if you start at high or low SoC. The likely cause is that something in the battery or cabling gets too hot and that the cooling system is not able to keep it cool enough. Too bad on this fine piece of battery! Translating these charge curves into charging times from 10 kWh to 50 kWh in the battery we get: Old charging curve: 22 minutes 16 seconds The conclusion from this data is that unfortunately, the Model S/X100 were never designed for higher power charging than 120 kW, and the increased peak power is mostly a gimmick for these cars. Model S/X90, BTX4 Battery (90 kWh) The first small increase in battery size from the original S85, the BTX4 90 kWh battery, has never been a real success. First of all, it's actual capacity is around 82 kWh - pretty far from 90 - and there have been signs of it degrading faster than other batteries. There has also been reports of users who have received power limitations on supercharging after a certain number of lifetime supercharges. So, our guess would have been that it would not have received any charging power upgrade at all. However, some of the data we have show that there seems to be a small push up to 125 kW for some vehicles. If this holds, the charging times for the S/X90 would actually be slightly improved. Translating these charge curves into charging times from 10 kWh to 50 kWh in the battery we get: Old charging curve: 24 minutes 52 seconds New charging curve: 23 minutes 18 seconds But again, the BTX4 is a tricky battery. Let us see what more data shows. Model S/X75, BTX5 battery (75 kWh) The 75 kWh battery has traditionally been a bit of a slow charger (in the Tesla universe), and a charging power upgrade would be a nice addition. As can be seen in the graph below, the BTX5 actually seems to really enjoy the higher power charging, which could be expected given that the peak charging power is still below the Supercharger-as-designed 120 kW and that the car probably has the same cooling capacity as the S/X100. Translating these charge curves into charging times from 10 kWh to 50 kWh in the battery we get: Old charging curve: 29 minutes 22 seconds New charging curve: 27 minutes 58 seconds This is a welcome charging time improvement for the small size Model S/X battery, especially since these cars have to charge more often. And given that the car and V2 Superchargers have been designed from 120 kW from the beginning, this is likely actually going to work well. TL;DR Tesla's increased charging power looks great for the Model 3 and the Model S/X75 since these car models were designed to take higher charging power. The increase for the Model S/X90 and 100 mostly looks like a gimmick to stay ahead on competition - something Tesla does pretty well anyway! How to contribute data? If you own a Tesla and use ABRP, please do login with MyTesla in ABRP, and check the boxes "Share data with ABRP" and even better "36h background sharing", which allows ABRP to poll your car for up to 36 hours after you have used ABRP. This helps us gather all your supercharging along a trip even if you don't use ABRP all the time. The MyTesla token which we get from Tesla after you log in is never stored permanently in our servers, only in your own browser. We definitively don't want to put your vehicle at any risk even if our servers get hacked.
  12. I am using N-Y since that also conveys other charger speeds than J-M (you have 100 kW, for example). But the parsing is quite hard-coded, so please don't remove any columns without warning me, it will break the import. Any updates will overwrite, and yes, other databases typically take priority so far.
  13. It turns out that the access for vehicles higher than 1.85 meters to the Euroshuttle line was somehow disabled in OSM. I added an access road, and also lowered the assumed vehicle height in our routing so that it passes the 1.85 meter limitation. Most EVs will be fine with this for now. The changes should be available within 24h or so on ABRP.
  14. Sorry for the delay, I just added an import function for your chart. Please check if data is what it should be and if we could add any more relevant info! The import happens daily around 23:00 UTC.
  15. Bo (ABRP)

    Delete ABRP account

    We are still missing that functionality, so just send an email to contact@abetterrouteplanner.com and we will do it manually for you.
  16. Thanks for all the input, just released this feature!
  17. This should now be fixed, thank you for the report!
  18. Since rollout of this has been a bit slow, we are still waiting for more reliable data. But yes, of course we will add this as well to ABRP.
  19. The tunnel works from France to the UK but not the other way around - some edit in OpenStreetMap the last week as obviously made the southbound direction fail. It would be most grateful if someone finds the problem and fixes it! (We run new map imports weekly, but let me know if you find it and I will re-import immediately) Edit: It works in the OSRM demo https://map.project-osrm.org/ so maybe it has already been fixed. I'll kick of a new map import and we'll see.
  20. I guess it should not say "Car" at all if you don't have temperature as part of live data. It will fall back to 20 C if nothing else is available.
  21. It seems to work for me also for 55%: https://abetterrouteplanner.com/?plan_uuid=3b2f320c-a83a-4e6e-90fb-778903cb5715 It takes one route there and a different route home.
  22. Thank you, now fixed!

Contact Us

Bo - Lead Developer and Tesla owner: bo@abetterrouteplanner.com

Jason - New Car Models, Developer and Bolt owner : jason@abetterrouteplanner.com

Idreams - Forums Administrator, Forums Developer and Tesla owner : idreams@abetterrouteplanner.com

×
×
  • Create New...