For those owners unhappy with drop in maximum charged range

[FlasherZ]

Both. If one keeps their car plugged in, and charges to at least 90%, they won't be posting about max rated range numbers below 260. Do you really think it's coincidence that both roadster and Model s owners that do the above have little to no range loss?

For what it's worth, prior to my battery pack going to the great recycler in the sky, I had 20,000 miles and a full range charge would only reach 257-258 miles after I followed rebalancing 3 times. I charged daily to 90% with the occasional range charge as necessary. The range charge would show only 252 miles until I followed rebalancing instructions and gained 6 back after 3 balancing cycles.

So yes, I kept my car plugged in, and I had < 260 miles. Now I'm back to 268 with a refurb pack after replacement.

 

[qwk]

For what it's worth, prior to my battery pack going to the great recycler in the sky, I had 20,000 miles and a full range charge would only reach 257-258 miles after I followed rebalancing 3 times. I charged daily to 90% with the occasional range charge as necessary. The range charge would show only 252 miles until I followed rebalancing instructions and gained 6 back after 3 balancing cycles.

So yes, I kept my car plugged in, and I had < 260 miles. Now I'm back to 268 with a refurb pack after replacement.

"A" packs throw a monkey wrench into the equation, since they have different firmware and behave differently. Since you no longer have an "A" pack, you will have to let us know how your new pack behaves assuming the same charging regimen.

 

[dennis]

The car will drift out of balance if you set the slider to anything less than 90%, or unplug it right away after a charge is finished(even if at 90%). There is nothing wrong with an out of balance pack, but getting it back into balance takes some effort and time. This is nothing new, and has been discussed many times in roadster threads.

Given that charging to less than 90% causes the battery to become unbalanced, why did Tesla replace the 93% "daily" charge control with the slider that allows you to charge to less than 90%?

 

[Super Gizmo]

^This is me with both Roadster and MS. Plug-in both cars every night to standard charge level with very few exceptions, regardless of miles driven. Each car gets range charged about 3-4 times per year as and when I need the safety factor. No complaints from me about degradation.

qwk & NigelM - Prior to the update bringing lower charge points and the capability to charge at a later time in the Standard Charge mode I always did what you have been doing and had a much batter range but what I can't understand is why Tesla encouraged people to charge at the lower charge points after that update. The update notes and comments by the Tesla people suggested that charging to a lower level than the full Standard charge was a better option than charging to the prior full Standard charge mode daily.

So, I and many other owners did exactly what we were encouraged to do and now have a much lower range. That is very disappointing. I know that the people at Tesla are busy with their expansion plans and higher production volume but they need to communicate to us the current owners exactly what is the right way to maintain and charge our batteries. There are too many conflicting suggestions on what to do and that is causing a lot of confusion and consternation.

 

[jerry33]

Given that charging to less than 90% causes the battery to become unbalanced, why did Tesla replace the 93% "daily" charge control with the slider that allows you to charge to less than 90%?

One reason was so that there could be a storage level. The other, and probably main reason, was so that the EPA numbers would stay the same. If they kept the previous settings, the EPA would calculate the range as the average between standard and range charge. The slider eliminates government stupidity (or more likely political pressure to stem the tide of EV vehicles by doing everything possible to make them look bad).

 

[djp]

Given that charging to less than 90% causes the battery to become unbalanced, why did Tesla replace the 93% "daily" charge control with the slider that allows you to charge to less than 90%?

Mostly because owners were asking for it, and Tesla engineers knew that dropping the SOC would extend the long term life of the battery.

They probably expected owners to do enough long distance trips to keep the pack balanced, or they screwed up the balancing/calibration algorithm at low SOC and are still working to fix it. I'm guessing the latter.

 

[markb]

But why do new cars still show 265 or greater miles at 100% even with the latest firmware? I don't understand how "new software" can make old cars have lower range (or hidden below zero, however you want to call it) but not new cars.

Exactly. But no matter how many times you or I or others make this point, that zombie theory refuses to stay dead.

Interesting. So you are confirming that the issues with 40 owners loosing range is mostly a balance and calculation problem. .

Might be just semantics, but I am uncomfortable with you saying I am confirming anything. From my perspective, only Tesla can confirm any theory regarding range loss being related to battery balancing issues, and as far as I know, they have not said a word about that, or even anything at all in regard to balancing in the S. I don't view the wildly inconsistent information received from Ownership or service centers to be reliable regarding this issue.

I just reported one owner's experience, and expressed my opinions, which may or may not be worth anything.

...The other, and probably main reason, was so that the EPA numbers would stay the same. If they kept the previous settings, the EPA would calculate the range as the average between standard and range charge. The slider eliminates government stupidity (or more likely political pressure to stem the tide of EV vehicles by doing everything possible to make them look bad).

Really? My understanding is that the epa fully charged an S, then drove it under varying road conditions, and determined how many miles they could drive. Had nothing whatsoever to do with Tesla's recommendations regarding charging.

 

[Gizmotoy]

Mostly because owners were asking for it, and Tesla engineers knew that dropping the SOC would extend the long term life of the battery.

They probably expected owners to do enough long distance trips to keep the pack balanced, or they screwed up the balancing/calibration algorithm at low SOC and are still working to fix it. I'm guessing the latter.

This is one reason why the issue is immensely confusing to me. Tesla's official stance is leave it plugged in and charge in the daily range zone, but apparently that's only partially correct (90% is the only setting that's good). There's the SC's recommendation to extend battery life (battery is happiest when operating between 60-80% so charge somewhere in there), and there's forum user opinions.

On any random commuting day I use 30-35 miles at most. How should I be charging? If I charge to 75%, perhaps I unbalance the pack but extend the life of the battery? If I charge to 90% I keep my pack balanced but reduce battery life? As far as I can tell there's no consensus. If I can't charge every night (a true statement, in my case), what then?

Indeed, I also wonder why 90% is some kind of magic number. On the LiIon packs my company produced, charging to 100% balanced the cells. By all accounts, the same top balancing is true for the Tesla. So how does charging to 90% fit in? How can you top balance cells you're not charging to the top?

 

[stopcrazypp]

Really? My understanding is that the epa fully charged an S, then drove it under varying road conditions, and determined how many miles they could drive. Had nothing whatsoever to do with Tesla's recommendations regarding charging.

If you have two distinct charging modes, (say 90% and 100%) the EPA will average your two modes and have your range rating be that. That's what happened to the RAV4 EV (rated 103 even though it gets 113 with 100% charge) and 2013 Leaf (rated 75, even though it gets 84 with 100% charge)

So instead of 265 EPA miles, the Model S would have been rated 252 (95%). By doing a update to eliminate the "Standard" charging mode shortly after EPA announced this policy, Tesla avoided having to change the range number on their EPA sticker.

- - - Updated - - -

Indeed, I also wonder why 90% is some kind of magic number. On the LiIon packs my company produced, charging to 100% balanced the cells. By all accounts, the same top balancing is true for the Tesla. So how does charging to 90% fit in? How can you top balance cells you're not charging to the top?

90% is not top balancing, it's closer to middle balancing (but not quite). It's still possible to do, but requires an accurate voltage to SOC mapping. But this point is kind of moot for the Model S as it appears NOT to balance at 90%. Only 100% has been confirmed to work. So charging to 90% does not really help the pack be in balance.

The Roadster on the other hand, DOES balance the pack even in Standard charging mode.

 

[Gizmotoy]

90% is not top balancing, it's closer to middle balancing (but not quite). It's still possible to do, but requires an accurate voltage to SOC mapping. But this point is kind of moot for the Model S as it appears NOT to balance at 90%. Only 100% has been confirmed to work. So charging to 90% does not really help the pack be in balance.

The Roadster on the other hand, DOES balance the pack even in Standard charging mode.

Yes, this is exactly what I meant, and you said it much better than I. Thanks.

 

[brianman]

Here is, unfortunately, a big part of the problem (not you Chad, but the way hidden loss works). It's going to be tremendously difficult to hit mass market if the range isn't considered trustworthy. Either not to degrade badly or to accurately display the range. If it says I've got 240 at max charge, I'm not leaving for a 240 mile trip saying "well, some miles are secretly hidden, so I'm fine" (all other factors assumed favorable for achieving the estimated range). Folks also aren't going to want to play some game of max charge, run it out, repeat for a week before a road trip so they have full range when they do leave. Or even as a "fix it once very 3 months" routine. That's simply not going to fly in the mass market.

A portion of the hyper aware and educated folks that visit this forum are a bit grumbly. Whether justified grumbles or not, that's going to be magnified 100x in mass market.

There needs to be some long term fix. Some way the charging software can rotate secretly emptying and filling some portion of the battery on overnight charges or some such. It must be a solvable problem behind the scenes at some point. After all, we're already discussing solutions here that, as a software guy, I would firmly call hacks to get things in balance. First hack it, then get it right?

Random observation:
This reminds me of Trim on SSDs. A lot.

- - - Updated - - -

Maybe a simple percentage with 'estimated 200-225 miles' for example that would change based on weather , elevation , speed...etc would work.

Offering kWh and/or SOC % in the UI as a setting rather than "Rated" or "Imaginary" would be a significant improvement.

I've never have an issue with my ICE vehicles "deceiving" me about range remaining because 1/4 tank is never "off". Units like kWh and SOC % would avoid this problem quite well.

- - - Updated - - -

If you are going to compare numbers, at least compare apples with apples. The Standard charge went from 93% to 90%. 245 rated was the 93%.

237 to 214 is a 9.7% drop. That's pretty significant, IMO.

- - - Updated - - -

(a) No, I don't have a similar loss in my car. You know why? Because my pack is balanced. (b) If the Model S truly did have a problem with degradation, everyone here would have similar losses. (c) Because they don't tells us that some owners follow simple instructions, and some don't.

(a) Wrong. You don't know if your pack is balanced. You also don't know if there's a causal relationship here. And, from a statistics standpoint, one data point is laughable.
(b) Wrong. Sometimes problems are the variability. My car had the 12V replaced preemptively. I never had a problem with it prior to that. Does that mean there was no 12V battery problem with the Model S?
(c) Wrong. On multiple levels. For now I'll just keep it simple and polite and focus on one here: at best you're being rude.

- - - Updated - - -

Isn't the point of "Standard Mode" in the Tesla to charge up to 90 percent? At what point are you in "Range Mode"?

Terminology clarification:

"Early" Model S firmware:
Standard charge = 92-93% (REST data gave us the specifics, no official number was published AFAIK)
Range charge = 100%

"Later" Model S firmware:
Daily charge = 50-90%
Trip charge = 100%

Note that I use "Range charge" above instead of simply "Range" to avoid confusion with "Range (driving) mode". This confusion may be part of the reason why Tesla changed the name to "Trip". Also note that "full charge" is an often confusing term so I avoid using it.

- - - Updated - - -

Yes 5.6 changed the way range is calculated. I noticed a ~10 mile drop in mine when it happened. I assume this is to 'zero miles' buffer.

There was a buffer below "0 rated miles" long before 5.6, or are you talking about something else?

 

[ElSupreme]

There was a buffer below "0 rated miles" long before 5.6, or are you talking about something else?

I am referring to the official change log notes from Tesla. I lost a good chunk of 'indicated range' with that update. I thought it was due to more 'zero mile' buffer than before. But I am unsure. I do remember having less indicated range after the update though. The note at the bottom heavily supports that 'indicated range' will be less, as it is trying to tell people they aren't actually losing any range.

Firmware 5.6

 

[brianman]

I am referring to the official change log notes from Tesla. I lost a good chunk of 'indicated range' with that update. I thought it was due to more 'zero mile' buffer than before. But I am unsure. I do remember having less indicated range after the update though. The note at the bottom heavily supports that 'indicated range' will be less, as it is trying to tell people they aren't actually losing any range.

I don't interpret anything related to "buffer below zero" in that screenshot. When I read them initially (and now as well, I guess), I read it as "heuristic for measuring and doing math changed" not "moved a chunk of the measurement below the zero line".

 

[ElSupreme]

I don't interpret anything related to "buffer below zero" in that screenshot. When I read them initially (and now as well, I guess), I read it as "heuristic for measuring and doing math changed" not "moved a chunk of the measurement below the zero line".

In both of my posts I state that I think it is due to more 'buffer below zero'. I never stated that Tesla said that. I did lose indicated range with the 5.6 update, it happened right before my road trip to SC. And my 100% charge went from ~265 to ~255 miles with the update. And my first leg was 250 miles exactly. It made me nervous to say the least.

And as for "heuristic for measuring and doing math changed" that statement is fully inline with your following one "moved a chunk of the measurement below the zero line".

Lets use an overly simple linear equation for demonstration.

Indicated Miles = y
Wh in pack = x

Start with this: y = 1/300 * x -2
Change to this: y / 1/300 * x - 10


This satisfies both of your statements above.

I did notice they changed the way distances are displayed during cold (sub 40F) temperatures. With 5.6, but I also noticed across the board loss in 'indicated range' that seemed consistent with changing the zero point. Either way it doesn't matter. Comparing range before 5.6 and after 5.6 is not really possible.

 

[brianman]

(a) In both of my posts I state that I think it is due to more 'buffer below zero'. I never stated that Tesla said that.
...

(b) And as for "heuristic for measuring and doing math changed" that statement is fully inline with your following one "moved a chunk of the measurement below the zero line".

(a) Right. I wasn't suggesting otherwise, at least not intentionally.
(b) I wasn't saying they're inconsistent, just that I didn't read the latter into what they stated. "I flew a kite today" and "it was sunny today" are potentially consistent, but the former doesn't necessarily include or imply the latter -- especially if you're a tester.

 

[qwk]

90% is not top balancing, it's closer to middle balancing (but not quite). It's still possible to do, but requires an accurate voltage to SOC mapping. But this point is kind of moot for the Model S as it appears NOT to balance at 90%. Only 100% has been confirmed to work. So charging to 90% does not really help the pack be in balance.

The Roadster on the other hand, DOES balance the pack even in Standard charging mode.

This is 100% incorrect. Both the Roadster and Model S balance at 90%. No range charging is needed.

 

[Andrew]

[I just posted in this in the "Decreasing Rated Range" topic as well, but since I take this as official word from Tesla, I thought it was worth repeating. Mods, please don't slap me.]

Our car is in for our first annual service today. I asked my service manager about the decreasing rated range. He said our battery checked out just fine, but to restore the rated range, here are the EXACT steps we should follow. I walked through the steps with him on the phone twice to confirm. This is directly from my service manager at the Santa Monica service center:

1. Charge to 100%. Let charging cycle complete.
2. Unplug.
3. Let the car sit for 20-30 minutes, up to a few hours.
4. Drive the car down to 20% SOC. This doesn't have to be immediately, it can be over a few days of use.
5. Once 20% SOC is reached, let the car sit for at least 20-30 minutes.
6. Plug back in and charge to 90% (or 100% if you need the range).
7. Repeat every few months, unless you charge to 100% regularly otherwise.

That's it! Done and done.

 

[stopcrazypp]

This is 100% incorrect. Both the Roadster and Model S balance at 90%. No range charging is needed.

Can you give a reference to a successful balance cycle initiated at 90% for the Model S? All the ones I have seen so far require charging to 100%, including instructions from service managers.
Charging to 90% seems to make little to no difference in the balance of the pack (to be clear, it might still do a little bit of balancing, but not enough to matter). There's still plenty of examples of people who charge to 90% regularly and still end up with lower range.

 

[MikeC]

[I just posted in this in the "Decreasing Rated Range" topic as well, but since I take this as official word from Tesla, I thought it was worth repeating. Mods, please don't slap me.]

Our car is in for our first annual service today. I asked my service manager about the decreasing rated range. He said our battery checked out just fine, but to restore the rated range, here are the EXACT steps we should follow. I walked through the steps with him on the phone twice to confirm. This is directly from my service manager at the Santa Monica service center:

1. Charge to 100%. Let charging cycle complete.
2. Unplug.
3. Let the car sit for 20-30 minutes, up to a few hours.
4. Drive the car down to 20% SOC. This doesn't have to be immediately, it can be over a few days of use.
5. Once 20% SOC is reached, let the car sit for at least 20-30 minutes.
6. Plug back in and charge to 90% (or 100% if you need the range).
7. Repeat every few months, unless you charge to 100% regularly otherwise.

That's it! Done and done.

I agree with this, it's consistent with the report from the field engineer that I've also posted in other threads:

The algorithm that evaluates the HV battery capacity is a feedback loop and it will take a few weeks to have it completely set to the customer's habit. I would recommend charging to 100% once a month and driving down to below 20% following the full charge. At 100%, the vehicle should seat, unplugged for 30 minutes and same thing at 20% prior to charging it. This is usually not necessary nor recommended but our algorithm needs improvements in evaluating of HV battery capacity which is directly related to range displayed. It currently under shoots the actual capacity when the customer doesn't charge to max or doesn't drive to below 50% (which is normally not the best for the life of the HV battery). Until we fix our algorithm, the customer should perform a deep cycle once in a while with rest periods at each end (30 minutes, not driving, not charging).

 

[qwk]

Can you give a reference to a successful balance cycle initiated at 90% for the Model S? All the ones I have seen so far require charging to 100%, including instructions from service managers.
Charging to 90% seems to make little to no difference in the balance of the pack (to be clear, it might still do a little bit of balancing, but not enough to matter). There's still plenty of examples of people who charge to 90% regularly and still end up with lower range.

There is documentation from Tesla for the roadster, but for the Model S, Tesla has just stuck with "plug the car in whenever it's not in use".

If you watch the Model S charging near the end, you can see the balancing occurring. With an unlocked car, the charge port goes from blinking to solid green, and after about 5-15 minutes the charge port goes dark. The instrument cluster stays on, and the voltmeter(on dash) also pulses during this time. After the balancing is complete the dash goes dark. The entire process takes about 30 minutes, which is the amount of time that Tesla also recommends the roadster be left plugged in after charge completion.