Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health

[AlanSubie4Life]

Today, with 4000 miles on the odometer, I have 72.8 kWh.

I have a LR AWD built in 9/19

Wow, that's a pretty brisk loss. Here I am at 13k miles and 13 months, and I have 300 rated miles which would be 73.5kWh with SMT.

I'd expect your 72.8kWh to work out to an actual 297 at a full charge...so perhaps it is a little higher than the 295 you quote? When did you last charge to 100%, or where did that 295 come from?

 

[golfleep]

What's your charging regimen?

Usually from 60% to 80% daily, or 40% to 90% over the weekends, I've done from 20 to 90% three times to try and see if the values would change (which it did not). Supercharged 4 times total, each time to 80%.

 

[golfleep]

Wow, that's a pretty brisk loss. Here I am at 13k miles and 13 months, and I have 300 rated miles which would be 73.5kWh with SMT.

I'd expect your 72.8kWh to work out to an actual 297 at a full charge...so perhaps it is a little higher than the 295 you quote? When did you last charge to 100%, or where did that 295 come from?

I have never charged to 100%. The 295 ( it is actually 295.56) comes from TeslaFi, which I understand is an extrapolated value.

 

[AlanSubie4Life]

Usually from 60% to 80% daily, or 40% to 90% over the weekends, I've done from 20 to 90% three times to try and see if the values would change (which it did not). Supercharged 4 times total, each time to 80%.

They can't all be gems. I wouldn't expect much to change this result. You're well within the normal distribution (though a bit on the early side). You can of course charge to 100% if you want but also would not expect much from that (maybe a mile or two at most) - clearly you don't have to worry about the possible impact of that on your battery available capacity loss!

I've seen claims here that when you charge above 90% that's when balancing takes place, but I haven't seen a reliable source, and it really doesn't make any sense, since it is important that the cells maintain balance all the time. Balancing isn't something that can just be put off indefinitely.

 

[golfleep]

They can't all be gems. I wouldn't expect much to change this result. You're well within the normal distribution (though a bit on the early side). You can of course charge to 100% if you want but also would not expect much from that (maybe a mile or two at most) - clearly you don't have to worry about the possible impact of that on your battery available capacity loss!

I've seen claims here that when you charge above 90% that's when balancing takes place, but I haven't seen a reliable source, and it really doesn't make any sense, since it is important that the cells maintain balance all the time. Balancing isn't something that can just be put off indefinitely.

Thanks for the reassurance! It's just a bummer knowing that I'm probably on the lower half (or quartile!) of the distribution, rather than the upper.

As for balancing - I have seen what appears to be balancing take place watching the cell max/min values in SMT, as the cell difference value closes as the car sits idling. For instance, it'll start off at 6-8 mV, and close to 2 mV a few hours later when look at SMT again. I've noticed it at all charge levels, so I think some degree of balancing is always taking place (assuming the values from SMT are accurate). Most of the time, my car is sitting with a cell difference of 2-4 mV.

 

[AlanSubie4Life]

It's just a bummer knowing that I'm probably on the lower half (or quartile!) of the distribution, rather than the upper.

For sure. It's possible you're just taking the hit early though. I guess we can always hope.

I'm interested to hear from the person who thinks they have the MOST rated miles at 100% after more than 20k miles. Based on the distributions posted above, and eliminating the inaccuracy spread, I'd be really surprised if anyone reports more than 74.5kWh after 20k miles (so that would be 304 miles for AWD, 318 for LR RWD, and different values for the 2020 vehicles (currently only the 3Ps are different I think)).

 

[leeeeee]

I have a LR AWD built in 9/19, and have had seen my rated miles drop pretty steadily since getting the car, and am currently sitting around 295 miles of rated range at 100%, with 4000 miles on the odometer.

I have near identical results. Delivered same week. 297 mi extrapolated range. 3300 odometer. I have no garage and NJ weather though.

 

[SleeperService]

30672 miles. 300 rmi extrapolated from 80% rmi of 240. Took delivery in September 2018.

 

[SD_Engnr]

15,475 miles. 272 mi extrapolated from 90% of 245 mi. Took delivery in August 2018.

It was a steady drop until December, when it dipped 14 miles, and it has since dipped another 3 miles. I reach out to Tesla every 4-6 months for diagnosis, and they always say, "It's normal". I find a 12% loss of capacity in just over 1 year to be anything but normal.

Here's their latest response:

It's worth noting that I never charge to between 60-80%. For the majority of 2019, I charged daily to 90% based on Tesla's recommendation, and I saw zero change.

 

[KenC]

15,475 miles. 272 mi extrapolated from 90% of 245 mi. Took delivery in August 2018.

It was a steady drop until December, when it dipped 14 miles, and it has since dipped another 3 miles. I reach out to Tesla every 4-6 months for diagnosis, and they always say, "It's normal". I find a 12% loss of capacity in just over 1 year to be anything but normal.

Here's their latest response:

View attachment 498621

It's worth noting that I never charge to between 60-80%. For the majority of 2019, I charged daily to 90% based on Tesla's recommendation, and I saw zero change.

Interesting. I used to charge to 80%, but as an experiment I now charge to 60%. After a full year, my LR-AWD shows 310 miles. Today, it was 312 miles, and yesterday 285 miles, because the battery was partly frozen.

 

[SleeperService]

Incidentally, at ~20% battery it projects to exactly 310 rmi on my car (I don't have the exact numbers, I neglected to take pictures at the time). This combined with @KenC 's anecdote above, it seems possible that the BMS loses the "bottom" of the scale more so (or at least in addition to) the "top" of the scale when kept between 40-80% (my normal range).

 

[crackers8199]

LR RWD, 28,310 miles, took delivery 7/31/18. just updated to 40.50.7 yesterday.

i normally charge to 80% because i drive very little (i work from home, so the majority of my driving is just running errands and such at night), maybe 30-40 miles a day if that on average. longer trips on weekends. we're leaving for a road trip this afternoon, so i charged to 100% last night...and the car is only showing 294. is this normal or should i schedule service? so many numbers in this thread i can't figure out what's normal and what isn't...

 

[AlanSubie4Life]

Incidentally, at ~20% battery it projects to exactly 310 rmi on my car (I don't have the exact numbers, I neglected to take pictures at the time). This combined with @KenC 's anecdote above, it seems possible that the BMS loses the "bottom" of the scale more so (or at least in addition to) the "top" of the scale when kept between 40-80% (my normal range).

I think it's more likely rounding error. Rounding errors are a lot more significant at a low SoC.

For example, let's say you had 62 rated miles at 20%. This extrapolates to 310 rmi. But that could actually be 61.55 rated miles at 20.49%. That would be 300 rated miles extrapolated.

You're less sensitive to rounding error in extrapolation at higher SoC. But the minimum error is about +/-1.5 rated miles (assuming you wait for a click up/down in the rated miles to get the precision there).

I think if you're really careful you can detect transitions on both % and rated miles and get even more accurate (and I kind of wonder why the apps don't use an algorithm for it - they already have the decimal precision on the rated miles from the API from what I understand; they would just have to play games to detect the SoC % transition).

 

[SleeperService]

Darnit @AlanSubie4Life quit raining on my parade, I want to think that I still have 100% of my battery capacity and I'll use MATH to make me believe it!

Seriously though, I think you are correct. As is, I'm pretty okay with my 300 rmi projected from 80/90% charge as that represents a 3.5% degradation over 1.25 years/31k miles. Should slow down from here. Keepin' calm and carryin' on!

 

[AlanSubie4Life]

As is, I'm pretty okay with my 300 rmi projected from 80/90% charge as that represents a 3.5% degradation over 1.25 years/31k miles. Should slow down from here. Keepin' calm and carryin' on!

Yes, this is one of the better results I've seen. Based on initial readbacks of vehicles showing 77-78kWh, it's more like 5.7% degradation, though it is the 3.5% degradation in the standard "76kWh max" framework (I'm working to figure out exactly how things behave in the transition), but it's still very good. @TimothyHW3 claims about 75.2 kWh at 20k miles (about 3.6% degradation, or 1.1% from 310 rated miles) but his car is less than 8 months old, so we'll see how it is after a full year. Mine took a dump around 7-8 months, so I feel like in addition to the mileage we also have to give these batteries some time to really waste away.

Can anyone beat 300 rated miles (more specifically, 73.5kWh, which is 314 rated miles for a RWD) at 30k or similar, with a vehicle older than 10 months?

EDIT: Looks like @KenC is a candidate...noticed that after writing the above. I'd want to see a result extrapolated from 80-90% though...but still seems like it will be good. Must be that cool Maine climate.

 

[SleeperService]

@AlanSubie4Life Mine went from 310 rmi to 300 rmi literally overnight in about that same timeframe (7-8 months old). I'm always suspicious of a step change like that... Again, don't have data or records to back that up, but I distinctly remember it.

 

[TimothyHW3]

Can anyone beat 300 rated miles (more specifically, 73.5kWh, which is 314 rated miles for a RWD) at 30k or similar, with a vehicle older than 10 months?

I have seen a SMT screenshot on another forum, where a guy had about 25-30k+ miles and still over 75kWh nominal full! This is almost 3% degradation after that many miles and almost zero lost rated kms.

I have also seen 70kWh at 15000miles. Not sure, but it seems either the model 3 battery is prone to degradation on some cars or BMS decalibration or both. Or maybe temperature and charging habbits. Don't know, but there are too many differences between packs.

 

[AlanSubie4Life]

I have seen a SMT screenshot on another forum, where a guy had about 25-30k+ miles and still over 75kWh nominal full!

POIDH!

Just kidding, I believe it.

Don't know, but there are too many differences between packs.

Could be those things. Manufacturing variation is also inevitable. People (and Tesla) blame the estimation for being off but my experience has been that the estimate of energy available, based on the trip meter result, is always pretty good (for me). A lower rated miles number means you're not going to get as much energy counted on that trip meter before you have to recharge, in my experience. May not be exact but correlates very well.

 

[KenC]

Not sure what you're looking for, but I've posted this before, but today, I've got yesterday's 284 rated mile dot all the way in the lower right corner:

As I've written before, that's one year's worth of data I pulled from Stats. Early January to early January. The estimated Rated Range is in blue; and the Temperature data is in red. I've added moving average lines to show the trends. Notice how well the moving averages correlate between temp and the estimated Rated Range.

Also notice, I have a blue dot at 284 miles, from yesterday. The car was frozen as the temp was about 16F and the wind was howling. My car sits outside. The Tesla app, showed a snowflake. Also notice last winter on the left, I also had 5 days with estimated Rated Range below 293 miles. I didn't know enough back then to check for snowflakes, but presumably my battery was frozen.

Below, is yesterday's Stats image showing my Rated Range is 284 miles, circled in red. The Battery SOC is showing at 62%, but that's not right. The car is showing 57%, and has a snowflake, meaning that even less is available.

Here, you can see both the Tesla app and Stats app. Stats is showing 62%, and the Tesla app is showing 57% with a blue snowflake.

And here's the Stats app showing the 284 mile dot:

What I conclude is that 3rd-party apps like Stats are pulling an API that is showing a number different than what the car itself is using. When I look at the car's app, and divide the odometer figure by the SOC%, I get between 308 miles and 312 miles, every day. Today was 312 miles. Stats and other 3rd-party apps are pulling a different SOC number. One that is higher than the one the car is using. That's why Stats is showing 284 miles in the above chart.

If you look at the next-to-last image where Stats shows Rated Range of 176 miles, divide that by SOC% of 62, and you get 284 miles. That's what it thinks is the Rated Range and it puts a dot there in my chart. BUT, if you use the SOC% that the car shows, 57%, and divide that into 176 miles, then I get 309 miles. Normal.

Someone told me once that there are two SOC API numbers. It seems that the car and the Tesla app are using a different SOC API number than 3rd party apps like Stats. This inevitably leads to confusion. My rated range is not 284 miles. It's between 308 and 312 miles, but because of this SOC API difference, I get these outliers on very cold days. Today, while the car says my Rated Range is 312 miles, Stats is showing 307 miles. And, if you look at the SOC number they are using, it's different by 1%, Stats shows 59% and the Tesla app is showing 58%. That's what usually is showing, a 1% mismatch. Today, the temps were better, about 32F. When it's colder, the mismatch can be higher, obviously yesterday it was mismatched by 5%.

 

[AlanSubie4Life]

Thanks for posting that comparison - it all makes sense but gives me a better idea of what to expect with the snowflake. If I ever see it again. I hadn't noticed that % behavior in the Tesla app since I use miles, good to know. It's good to be able to see both at once.

and divide the odometer figure by the SOC%, I get between 308 miles and 312 miles, every day

Most of this is rounding error. It can be eliminated if you're super careful to watch the %, but there's really very little point.

Yep. Your battery is very healthy, and you have minimal degradation (looks like you might just be starting to "poke out" from the 310 mile point (very hard to say), in which case you're right around 76kWh vs. the ~78kWh you probably started with).

Nice. You do have relatively few miles so far though. That being said mine took a sudden dump at around 6-7k at ~7-8 months (a 4-6 mile overnight loss coincident with a software update). So you're doing well!

Also hasn't been quite as cold for you this year. I would expect for you to see slightly deeper swings on the blue line this winter (about 6-9 rated miles deeper) at a given temperature, due to hidden degradation (my new pet theory) becoming more visible. Arguably you can already see this with the first cold two days this year giving you the lowest value you have ever seen (lowest kWh) though more data will confirm it. Last year you had an extra 2+kWh or so available so it had to get much colder to bring you down to the same level. That’s the theory anyway. Obviously the temperature won't directly correlate here because it does depend on the soak time as well. Looks like the latest two datapoints were fairly well cold-soaked.