Evaluation and Observations with v6.1 Trip Energy Prediction

[jerry33]

There's a difference between range anxiety and inconvenience of refueling an ICE car. You can preplan stops in an ICE car too, if you want. Let's keep it real, you have to be extremely absent-minded to have range anxiety in an ICE.

You could, but no one does, hence the ICE range anxiety.

 

[Vger]

Since I don't have 6.1 yet, can you now enter a route with multiple waypoints? This is one feature that's missing and often needed for road trips.

As ecarfan points out, via points are STILL not there. This is even more problematic now, in a sense, because if range prediction is critical on a complex, lightly travelled or "scenic" route, you cannot get the prediction that you need.

I sure hope they get to this soon! It has been promised, as all this stuff has, so eventually it should pop out of the Tesla software mill. :wink:

- - - Updated - - -

Concerning range anxiety: I will respectfully and playfully put it this way: If you have not experienced range anxiety in an EV, you are just not trying hard enough!

Seriously, but that I mean that the real pioneering, adventuring and learning in EV ownership comes with pushing boundaries— extending trips beyond the supercharger network, seeking out scenic byways, refusing to take an ICE on a tougher trip, going when the weather is not ideal.

I have enjoyed doing all that in the >120,000 kms I have driven in our two Teslas since 2010. Have I screwed up a few times? Sure. But I am quite sure I will not be screwing up nearly so often anymore, with range prediction in place.

 

[Cottonwood]

My limited experience with the energy prediction seems to indicate that it is updating the future use based on a some recent average use, at least while you are driving. This means that speed and weather are taken into account implicitly because those speed and weather factors affect current energy use.

I really like the updates to the display while you are charging. As Vger say, that offers a very nice way to see when you have enough energy in the battery to head on to the next charging point. Still, I am still better calibrated on using EVTripPlanner than using the car's energy predictions, and I know exactly what I put in for temperature and speed adjustments.

 

[David99]

My limited experience with the energy prediction seems to indicate that it is updating the future use based on a some recent average use, at least while you are driving. This means that speed and weather are taken into account implicitly because those speed and weather factors affect current energy use.

So it does adjust to different driving styles! That's awesome!

 

[aviators99]

There's something else going on in the prediction. Here in South Florida there are no hills, so there are typically no bumps in the prediction line. But on Sunday (on my way back from the Tesla drag races) I noticed a little bump. I was very curious to see where that would be. It turned out to be where I exited one highway (Florida Turnpike) and entered another (Sawgrass Expressway). I always have quite a bit of regen during that period, and that's what it seemed to be taking into account, as my actual usage mirrored the prediction (from 60+ miles before).

 

[MarcG]

THIS IS A BIG DEAL: The prediction is re-calculated even during charging! So at a supercharger stop, if you put in your next charging destination, you will get a direct indication of when you have charged sufficiently to reach it— no mental math required! [The crowd roars!]

Thank you for the detailed report! Would you (or anyone else already on 6.1) be able to post a screenshot of the prediction recalc during charging?

 

[tomas]

What would be great for long trips: dynamic suggested charging waypoints. Let's say Chicago to Denver. Start in Chicago, put in trip to Denver. As you are nearing a SC and in need of charge, nav could ask if you want to go to SC, then - if you accept - navigate you to/fro, while still retaining your ultimate trip destination. Even better if it could alert you when you have charged enough to get to next SC without changing ultimate destination.

 

[RiverBrick]

Thanks to the OP for the detailed summary.

So far, there seems to be no evidence that it take into account weather conditions, especially wind. In fact, our experiences yesterday indicate that the prediction will be overoptimistic in head winds and pessimistic in tailwind conditions.

It needs more observation, but there's a chance it takes into account the current ambient temperature measured at the car. The system seems to get way more optimistic when it's launched in a heated garage. (As compared to the down to -25C the car has seen recently outdoors).

Also, from comments in the Official LA to NY videos, it appears the January 2014 rally vehicles had this feature. One year or more seems like a long time for beta-testing.

 

[Cottonwood]

Cross posted from the 6.1 main thread:

Here are some comparisons on the range prediction from .92 to .113:

I am in Boulder with both my early Sig P85 and the recent P85D. Both currently have v6.1(2.2.113) software as of this morning.

From 9 days ago, the P85D with v6.1(2.2.92) reported needing 43% charge to get to Silverthorne, CO and 24% back.

Today the P85D with v6.1(2.2.113) reported Silverthorne as 46% there and 28% back, and the P85 with v6.1(2.2.113) reported needing 41% to get to Silverthorne and 23% back.

It is interesting that the P85D estimates changed by 3-4% SoC over an 80 mile trip for .92 to .113. It's also interesting to see the P85D needing 10% over the 64% that the P85 needs for a 160 mile round trip. That is an increase in predicted energy usage of 10/60 or 16% for the P85D over the P85!

 

[Rlsarch]

Who wants to to see the prediction in Projected miles left at destination instead of percentage???

Tesla has calibrated me to managing to miles left and then they go and through percent of battery at us! Who know how far they can go an a quarter tank of gas? Samething, who knows how far they can go on a 25% battery?

 

[ljwobker]

It's possible that they're trying to push people towards using percentage as the main unit of discussion rather than miles... personally I would prefer kWH, much like in an ICE I would greatly prefer gallons/liters... but that's unlikely to sway anyone

 

[jvonbokel]

It's possible that they're trying to push people towards using percentage as the main unit of discussion rather than miles... personally I would prefer kWH, much like in an ICE I would greatly prefer gallons/liters... but that's unlikely to sway anyone

Agreed on the % vs kWh. One of my pet peeves is articles about charging stations that specify the time to achieve an "80% charge". 80% of what, exactly? Depends on the vehicle, obviously, but that's not so obvious to the uninitiated.

 

[bollar]

Had an interesting bug last night. (2.2.115)

I went to go survey the Ardmore, OK Supercharger site, which is a 185 mile round trip. Left with a 100% charge (Ideal: 294.1, Rated 254.2, Estimated: 228.1). Trip energy prediction said the round trip would leave ~16% battery and I drove at or 5 mph above the speed limit to stay close to that range. The bulk of the trip was 65 MPH aside from a 75 MPH stretch in North Texas and 70 MPH in Oklahoma and was consuming less energy than predicted -- I didn't look closely, but it was around 280 wh/mi.

Arrived in Ardmore with somewhere between 55-60% battery remaining, took some pictures Supercharger - Ardmore, OK - Page 2 and started heading back. Winds from the South had picked up during the drive to about 25 mph and consumption was much higher (~350 wh/mi). No worries, slowed down to about 65 mph and was able to stabilize the battery percentage estimate at destination to ~11-13%.

At the Texas rest area north of Gainesville, driver got out and car turned off. When she got back in and the screen turned back on, the estimate had dropped to 2%!

We continued our drive south, driving 60-65 MPH and kept the estimate between ~0-2%. Fortunately, there was a fair amount of right lane traffic doing 65 in the 75 zone, so we we weren't creating any danger. When we got off the ramp in Denton to check out the Denton Supercharger site, the battery percentage estimate jumped up to 11%.

We had a 30 minute stop in Denton and started heading home. On that drive, the estimate was fairly consistent at 11%, until we got within 10 miles of home, at which point, it started jumping between 11% and -2%! By jumping, I mean the screen would show 11% and at the next refresh it would be -2% and a refresh beyond that would be back at 11%. This started happening about the same time power limiting began (about 15%, I guess). We arrived home with (Ideal: 27.3, Rated 23.6, Estimated: 19.3) miles.

Also temperatures were in the high 60s, so I wouldn't expect battery temperature to have been an issue.

 

[Lump]

Trip estimator graph is great, kids no longer ask "how much longer before we get there", they could see the graph & estimate arrival time on their own, downside is it gave them something else to argue over .

 

[cpa]

I recently drove round-trip from home to Modesto for a few days, then drove up to Manteca to see if the SC was operable (not!) then came home. I set the trip estimator upon departure from home and again upon leaving Manteca for home. There is a stretch of highway that had been realigned about 6 months ago, but is not recognized with the maps on the touchscreen, so it kept giving me crazy directions to get back onto the old, non-existent alignment. Each time the map would recalculate the trip estimate, and my residual would fluctuate by two-three percent, but not to the drastic extreme of Bollar. After a couple miles it would display the original state of charge at my destination. It was remarkably accurate, despite these goofy temporary fluctuations.

Since his aberrations occurred after stopping for short detours, I wonder if the program re-calculates the route estimate with spurious data, and somehow the display does not correlate to the calculation. Bollar mentions "refresh," which I am unfamiliar with. Did he cancel the trip, and then plan it again?

Perhaps this is another reason to use the trip planner as a guide but do not rely upon it as definitive. And, always, always have a reasonable buffer built in for all longer journeys!

 

[bob_p]

There appears to be a bug in the range estimation software (I'm still on .115). While driving on a trip last weekend, we had several times when the range jumped from a positive number to something negative. Cancelling and re-entering the destination corrected the problem.

Anyone encountering this situation should do a BUG REPORT before cancelling their destination, to give Tesla more examples of the problem.

 

[mknox]

I have not had much luck with the Energy Prediction app. It is always more optimistic compared to what I'm actually getting. This morning I tried really hard to come close. I started out with a fully warmed up battery and cabin. Charging completed about 45 minutes before I left and I had the HVAC on (remotely from the app) right up until I left. No power or re-gen limits at all. I drove at or below the speed limit. Much of this trip was on the freeway and for as much as I could, I drafted behind large trucks. I kept the cabin heat down and wore a large parka. I have ultra low rolling resistance tires, all properly inflated. The weather was sunny and the roads were clear and dry, but it was cold (-15 F when I left and - 4 F when I arrived).

I'm not sure what more I can do. If I drive "normally" and keep the cabin heat at a comfortable level, it looks much worse. Others seem to report being able to meet or even beat the prediction. Interested in any tips, trick or techniques.

 

[tomas]

Here's a tip: move south!

 

[mknox]

Here's a tip: move south!

For sure! I was supposed to be in SoCal this week, but my business meeting was cancelled. Having said that, I thought the prediction app was supposed to account for temperature.

 

[HankLloydRight]

mknox -- just curious, what was your average Wh/m for that trip?