Decreasing rated range.

[Kipernicus]

Stats at one year: 100% = 192, 90% = 169
Have done around 10-20 max charges, and supercharged many times (perhaps 40? didn't keep track)

 

[FredTMC]

Stats at one year: 100% = 192, 90% = 169
Have done around 10-20 max charges, and supercharged many times (perhaps 40? didn't keep track)

car's mileage? Thx!

 

[hans]

Stats at one year: 100% = 192, 90% = 169
Have done around 10-20 max charges, and supercharged many times (perhaps 40? didn't keep track)

Assuming you mean Rated Miles correct? Since rated miles have some component of driver history, and the algo for Rated miles has not been the same since the first deliveries, it's better to compare Ideal Miles (even though they are useless for just about everything else). At least that what my service advisor told me.

 

[bluetinc]

snip... Since rated miles have some component of driver history, ...

Rated miles have NO component of driver history. Anyone who has told you so is incorrect and should be corrected. Tesla engineering has confirmed this, as has all the data that everyone has collected. There have been some misinformed Tesla employees that have spread this misinformation, (which seems to keep popping up).

Peter

 

[hans]

Rated miles have NO component of driver history. Anyone who has told you so is incorrect and should be corrected. Tesla engineering has confirmed this, as has all the data that everyone has collected. There have been some misinformed Tesla employees that have spread this misinformation, (which seems to keep popping up).

Peter

I did challenge him on this and he promised to send me a email from engineering about it. Still waiting ;-)

Nevertheless, the Rated miles algo has not been consistent between all firmware releases so comparisons between different firmware releases can still be problematic.

 

[apacheguy]

Nevertheless, the Rated miles algo has not been consistent between all firmware releases so comparisons between different firmware releases can still be problematic.

Then how do you explain brand new cars with 5.8 still getting > 265 rated on a range charge? If we all compare our range charges I'm pretty certain that we're comparing apples to apples.

 

[ZBB]

Arizona here, lifetime usage is 293 Wh/m and dropping. Recent 100+ mile errand trips were done using 270 Wh/m or less. That's just how I drive.

Wait till your summer hits and you are forced to use the AC indiscriminately!

I'm also in AZ, and took delivery at the beginning of last summer. My lifetime Wh/m is at 297. Since it cooled off, I've been averaging right around 293.

Summer started off averaging ~308 the first 2 months -- and I only noticed heavier battery drain when cooling the car down after being parked outside during the day. That would last about 10 min, and then returned to close to normal after the interior cooled down.

We had a cold spell in early December... Lows below freezing, highs in the mid 40s. Energy use was worse then than summer -- I averaged about 315 Wh/m over the first week or so in Dec...

 

[bluetinc]

I
Nevertheless, the Rated miles algo has not been consistent between all firmware releases so comparisons between different firmware releases can still be problematic.

It's good to understand the differences that have happened along the way. So far, I think all of these changes effect ideal miles along with rated miles, except #1.

1. The "rated mile" energy unit can be different on different cars. I have seen 306 and 302 Wh/mi.
2. Temperature "corrections" have been updated over time to account for cold differently. This will show up at cold temperatures (sub 50F).
3. "Standard" charges use to mean 93% SOC, which became 90% and now with a slider gives anywhere from 50% -> 90% SOC.
4. Adjustments have been made to charge termination voltages for non 100% SOC charges to improve charge termination.
5. For the first time, with software 5.8, it "appears" that some capacity was moved to below 0 miles, (0 miles now equals 320V instead of 316V on a 85kWh A battery), but I can't confirm that this isn't part of the new temperature compensation until it warms up this spring.

I'm not sure why "Rated Miles" is thought to be less consistent than "ideal miles", but regardless none of these changes explain the 30+ mile drops in displayed miles owners have seen.

I now feel that Tesla Corporate would prefer to keep the FUD flying than really explain what's going on to those of us that show the largest drops and why it's so different between particular cars. Consistently suggesting that there are changes in mileage calculation and charging without details serve this well. Somehow, no matter what changes have taken place, if you walk over to a new car with the same software, they seem to show 267-275 "rated miles" just like my car did when new.

Peter

 

[hans]

Then how do you explain brand new cars with 5.8 still getting > 265 rated on a range charge? If we all compare our range charges I'm pretty certain that we're comparing apples to apples.

You are comparing apples to apples, and I can not explain it. That is exactly what I am asking Tesla to explain with my car. All I am saying is do not compare the rated miles you got with 4.x when you bought your car to the rated miles you get with 5.x now. If for no other reason other than that you are playing into their corporate FUD.

 

[gg_got_a_tesla]

On a visit to the Fremont SpC for Flat Stanley's sake, clocked 174 miles Rated at Standard today. Car's at 18,349 miles.

 

[gg_got_a_tesla]

Was talking to Kipernicus earlier today about the difference in range degradation for our 60s (I've done 50% more miles than him lifetime); I theorized that my car may have benefited a bit from being plugged into a 110V outlet for 8-9 hours on all weekdays while I'm at work?! Don't know if that theory has any legs...

 

[strengthcoach4]

Anybody else notice different range for different profiles? When charged under my wife's profile the car usually finishes with about 5 more rated miles compared to when charged under my profile.

 

[brianman]

Nevertheless, the Rated miles algo has not been consistent between all firmware releases so comparisons between different firmware releases can still be problematic.

Then how do you explain brand new cars with 5.8 still getting > 265 rated on a range charge? If we all compare our range charges I'm pretty certain that we're comparing apples to apples.

Firstly, he doesn't have to explain it. Secondly, many owners (myself included) on TMC have reported on numerous occasions that across a firmware transition the rated miles has decreased -- both instantly, and comparing "before week" and "after week" even with consistent weather. Now that that's been set straight...

To your question: One explanation is newer chemistry. Another explanation is that the rating algorithm has a time component -- regardless of actual cell capacity. There are many possibilities.

 

[Gear]

Firstly, he doesn't have to explain it. Secondly, many owners (myself included) on TMC have reported on numerous occasions that across a firmware transition the rated miles has decreased -- both instantly, and comparing "before week" and "after week" even with consistent weather. Now that that's been set straight...

To your question: One explanation is newer chemistry. Another explanation is that the rating algorithm has a time component -- regardless of actual cell capacity. There are many possibilities.

My guess is that it has somebody to do with balancing. Maybe the older firmware versions were more optimistic when the pack was imbalanced. Since the pack would be balanced when the car is brand new, it'd read the full 265 on either. Of course, as you said, it could be any number of things.

 

[apacheguy]

Firstly, he doesn't have to explain it. Secondly, many owners (myself included) on TMC have reported on numerous occasions that across a firmware transition the rated miles has decreased -- both instantly, and comparing "before week" and "after week" even with consistent weather. Now that that's been set straight...

Right, but what I was saying is that if we compare range charges on the same version (I.e. 5.8) then we are comparing apples to apples. We don't have to concern ourselves with the "improved" algorithms.

My guess is that it has somebody to do with balancing. Maybe the older firmware versions were more optimistic when the pack was imbalanced. Since the pack would be balanced when the car is brand new, it'd read the full 265 on either. Of course, as you said, it could be any number of things.

Maybe, but then this would beg the question as to why the newer firmwares don't simply correct the issue by balancing the pack.

 

[brianman]

Right, but what I was saying is that if we compare range charges on the same version (I.e. 5.8) then we are comparing apples to apples. We don't have to concern ourselves with the "improved" algorithms.

Maybe. But how long of a balancing period do you allow for? An hour? An hour for 3 weeks straight? Range charge every day starting last November? Unfortunately, it's not quite as simple (IMO) as you suggest.

 

[stopcrazypp]

Maybe, but then this would beg the question as to why the newer firmwares don't simply correct the issue by balancing the pack.

It's probably not a good idea to trigger a balance cycle with a firmware update. The car might not be charging while it's updating (or owner might unplug to soon for a balance cycle to finish), so it'll need a mechanism to pre-schedule a balance cycle for the next charge. And for the Model S, it seems it doesn't balance without doing a full charge, so you might also need to force a full charge (which the owner might not want to do).

 

[yobigd20]

many owners (myself included) on TMC have reported on numerous occasions that across a firmware transition the rated miles has decreased -- both instantly, and comparing "before week" and "after week" even with consistent weather.

That's because they changed the standard charge level %. "Standard" used to be 93% charge. Then with the new slider bar, what we are calling "standard" now (that little line) is 90%. So the jump from 245 "standard charge" to 231 "standard charge" is simple because I'm only charging to 90 percent now instead of 93%. Aside from that, I haven't noticed any degradation between software firmware updates. My range charge is still > 255 and i am at 36k miles.

 

[wycolo]

> how long of a balancing period do you allow for? An hour? An hour for 3 weeks straight? Range charge every day starting last November? Unfortunately, it's not quite as simple (IMO) as you suggest. [brianman]

Which is why I was surprised to hear that during a 'quick' balancing charge at a Service Center the owner was told that all strings were 'within one millivolt' of each other. I still can't believe it.
--

 

[nrcooled]

Anybody else notice different range for different profiles? When charged under my wife's profile the car usually finishes with about 5 more rated miles compared to when charged under my profile.

This is very interesting and I would like to know more. I am going to charge under my wife's profile tonight to see if it makes a difference.