Wiki Sudden Loss Of Range With 2019.16.x Software

[sdoorex]

Yup, repeatedly... advice from Elon is to just keep your car plugged in. The effect there is you immediately charge and stay sitting there at that charge until you drive next...

And the indicated range of "daily" charging includes up to 90% on the dial. It would be perfectly reasonable for a person to think setting to 90% is an acceptable every day occurrence thing to do.

In summary, the combined advice from Tesla amounts to "just leave the dial at 90 forever and plugin whenever you can".

I rarely charge above 90% and I used to only charge to 60% or 70% depending upon time of year when I had access to L2 charging because I use about 20% per day. After I moved to a place with just L1 charging, I increased my charge limit to 70% or 80% depending upon time of year, that is until Elon tweeted this and I further increased my charging to 80% or 90% depending on time of the year. I also supercharge only about 5% of all charging yet I still lost at least 22 miles of range due to the update.

Not worth going below 80% imo. Even 90% is still fine. Also, no issue going to 5% or lower SoC.

 

[scottm]

My model S was not dramatically affected by the malware... but I think my new realities are:

• I'm afraid to take any software update from Tesla until they come clean on this. Reject reject reject... and I hate v9 UI that I'm stuck on and desperately want something better.
• I'm afraid to charge beyond 90%. I'll never get to test my actual 100% range by charging all the way up.
• My new "full" is 90% and it would take an emergency for me to charge beyond that. I wonder how I'm going to do in winter?
• I guess I'm stopping at every supercharger now on the highway for trips. So that's more supercharge cycles. What's the worse evil, higher charge or more cycles?
• And I'm going to be spending more time at superchargers because it's slower charging than it used to be.. even though I'm "unaffected" by the reduction. Do I call them "Not So Superchargers" now? And this comes at a time when Tesla is making SuperDuperchargers for model 3's to enjoy.

 

[willow_hiller]

My car was not dramatically affected by the malware...

Ever heard of Hanlon's Razor?

"Never attribute to malice that which is adequately explained by stupidity."

Makes no sense for Tesla to deliberately hobble their flagship vehicles. Much more likely this is a poorly thought out bug fix that introduced even worse bugs.

 

[Droschke]

In summary, the combined advice from Tesla amounts to "just leave the dial at 90 forever and plugin whenever you can".

Absolutely. That's how my car charging routine has been.

In 4 and half years of ownership, the dial always set at 90%: schedule home HPWC charging (set at 24amp) to 90%, and ad hoc supercharging to 90%.

Twice have charged to 100%, once when the car was new and on a long trip to barely make it to the next supercharger and once in 2018.

Nevertheless, at 43k miles I lost 30 RM overnight (~12% loss). Not normal and not normal degradation.

 

[scottm]

Here's a factoid from an "unaffected range" car that I'm lucky enough (so far) to have... (he says while knocking on wood dash)

I plug my car in all day at 6 kw charging at work ... for 8 hours a day. And leave it unplugged at night at home. For almost all of the 5 years I've owned it..

This, along with any kind of convenience L2 I can find around the city, and many routine supercharger visits (1 a week).. And I have gone up to 100% on many occasions .. while road tripping, so I don't stay at that level for more than 10 minutes as it burns off quickly on the road again

I just wonder if the slow bake all day has anything to do with being lucky?

This is an original 2014 85 battery.

 

[Chaserr]

It depends on what your definition of 100% is, and that's the trap you fall into.

It would be best to wrap your mind around this: a full charge is less than 100% battery capacity.

That's the magic sauce. You literally do not know what the battery pack capacity is. You just get to know how much range you can expect from a fully charged battery. And that should be how the car is sold. Tesla is moving toward this model.

The EPA rated full charge at 4.2v. Reducing voltage at 100% is changing the EPA rating without the EPA's consent. Illegal.

If Tesla would software limit charging a brand new battery to a lower max voltage than that used when they did the EPA certification, then their EPA ratings would be false and the class action suit would have merit (and even the EPA would have skin in the game, cfr. the links posted some pages ago).

I haven't seen evidence that a brand new 85 kWh battery is being software limited in a way that makes the EPA rating invalid (other than perhaps the odd service error in failing to reset some failure conditions when a battery is swapped, IIRC). What they do to batteries that are not brand new is quite different, and apparently the range reductions don't happen to all older batteries either.

You hit the nail on the head here. Tesla is changing the EPA rating after sale. VW did this and was sued to high heaven, the news called it "dieselgate" and the punishments for post-sale EPA mileage adjustments in software are, precedence wise, extremely severe. Tesla is returning the voltage before an EPA investigation forces them to buy back cars like they made VW do. Changes to teh EPA's pre-sale ratings are taken seriously and it would be incredibly ironic for Tesla to be handed a dieselgate conviction. The EPA absolutely will not allow companies to cheat their official EPA tested numbers.

 

[mjmiron]

Think about how much time we all have used up on this battery issue trying to search for a resolution from a mute company? I think we will know an answer in the next two weeks either from my complaint with the Attorney General or the other folks that have used arbitration.

The online forums are an absolute godsend from the point of data and ideas shared. Thank you to all the folks with scan tools that have shared the invaluable data and those who shared their stories. Like i stated earlier we are the 2% pioneers fighting for the rights of future EV owners.

 

[David99]

Absolutely. That's how my car charging routine has been.

In 4 and half years of ownership, the dial always set at 90%: schedule home HPWC charging (set at 24amp) to 90%, and ad hoc supercharging to 90%.

Twice have charged to 100%, once when the car was new and on a long trip to barely make it to the next supercharger and once in 2018.

Nevertheless, at 43k miles I lost 30 RM overnight (~12% loss). Not normal and not normal degradation.

Ok charging to 100 % a lot can't be the trigger either. I have charged to 100 % probably 80 times over 5 years.
I think we really tried to every possible scenario or habit that would cause the issue but there really isn't any correlation we can find. I think at this point the only reasonable explanation is that the affected batteries have some defect that triggers the limit.

 

[MP3Mike]

Think about how much time we all have used up on this battery issue trying to search for a resolution from a mute company? I think we will know an answer in the next two weeks either from my complaint with the Attorney General or the other folks that have used arbitration.

You really think you are going to get any meaningful information from an initial response to the AG forwarding your request to arbitration? I can pretty much tell you their initial response will be a denial of most things and saying that they are already in the process of mitigating the range loss. (Look at the information they "share" in the arbitration cases regarding the yellow border on the MCU screen. Nothing of any value.)

The other thing to remember is once the lawsuit was filed that pretty much puts a gag on them sharing anything outside of the lawsuit. (Assuming that they would otherwise share any information.) So the lawsuit might actually delay/prevent you from finding anything out via the normal methods.

I would expect it to take many months, if not years, to get anywhere in the legal system.

 

[sixela]

How is the state of charge with corresponding voltage a proprietary secret? LOL

Less snippiness, please.

The BMS software is definitely a trade secret. How it tries to detect a chemical condition in some of the cells and the proper response to ensure safety and to ensure a longer life also are, since these details expose characteristics of the battery chemistry.

 

[sixela]

For what purpose?

for safety reasons, to ensure less degradation of the battery in future...you name it, I have only guesses, but "Tesla does it just to spite their customers" is not one of them. Tesla isn't exposing this, but there must be some reason for capping Vmax and there must be some trigger.

Obviously, the packs that are affected are now "below average", but as said before, not everyone can have above average batteries.

Also, the BMS only has limited data so it's perfectly possible for some false positives to occur (i.e. for the BMS to cap a battery when in fact, if you knew exactly the state of the entire battery pack, that would be unnecessary).

And I'd expect them to first err on the side of caution and avoid false negatives (a BMS that would not cap Vmax when it would be the right thing to do) as a primary consideration, only tweaking to avoid false positives later.

Unfortunately devising tests that have no false positives and no false negatives is not possible.

 

[mjmiron]

You really think you are going to get any meaningful information from an initial response to the AG forwarding your request to arbitration? I can pretty much tell you their initial response will be a denial of most things and saying that they are already in the process of mitigating the range loss. (Look at the information they "share" in the arbitration cases regarding the yellow border on the MCU screen. Nothing of any value.)

The other thing to remember is once the lawsuit was filed that pretty much puts a gag on them sharing anything outside of the lawsuit. (Assuming that they would otherwise share any information.) So the lawsuit might actually delay/prevent you from finding anything out via the normal methods.

I would expect it to take many months, if not years, to get anywhere in the legal system.

You may be right about them withholding information but i think your wrong about them sharing more excuses or a possible fix. I feel pretty strongly this will force Tesla to fix this before it goes to court especially since it will dive into proprietary information about the batteries.

 

[Droschke]

Now that there's a class action suit and legal gets involved it's not going to get any better. Yes, some information will be extracted (by pulling teeth) in discovery, but since it involves trade secrets not all of it will actually be publicly disclosed, I guess.

I hope the suit, at minimum, would reveal the hidden information displayed by the 3rd party apps (which I don't use, btw) and which Tesla is not making available to the owners, in addition to the real reason behind the capacity cap beyond the lame response that this is "to protect" the batteries.

Also, I might have missed it, and if so my apologies in advance, but would you state how miles have you lost?

 

[Chaserr]

If Tesla's german translated note is real, the reduction is "protecting" the batteries from needing warranty service. they're trying to save money by avoiding some condition that will necessitate them paying for battery repairs. That's the gist of what I got from the translation, at least. They have no reason to care about batteries out of warranty, but in warranty batteries are expensive and apparently those impacted by this downgrade need service.

I agree with David, it's likely triggered by an underlying defect. Tesla has given nothing and our own data shows it's too random to be any of the suggested causes. Also, defects are warranty covered and tesla is trying to avoid warranty coverage using the vmax limit to reduce whatever causes the defect to manifest.

@sixela We already know the limit isn't imposed by the BMS software - otherwise replacing the battery (where the BMS lives) would fix the issue. It doesn't. This has been verified on multiple owners' cars, the limit has to be removed with an MCU update like any other software cap tesla has used.

 

[sixela]

The BMS very likely has parameters and code that are set by the downloadable car firmware. If a software update changes the BMS behaviour by capping Vmax then it's not entirely in the battery pack.

What you may seen as a "defect" Tesla may see as a rare condition in a battery that nevertheless is seen as degradation uncovered by warranty, unfortunately. Given the lack of promises in the warranty with respect to long term battery capacity (and the disclaimer language) I'm not giving you a high chance of success to force Tesla to change its stance using litigation, even though there is quite evidently a huge customer satisfaction issue (and even anger).

Even on a Model 3 similar things could happen. Given that they do put a stick in the ground and guarantee 70% of the original capacity at the end of warranty, they might program the BMS to ensure that some batteries acquire charging profiles and limits that ensure that those 70% are indeed there at the end of warranty, and some users would be unlucky compared to some others.

You could, to give one example, not force them to give users unrestricted capability to run their Model 3 batteries into the ground just before the end of the warranty period just to get a new battery.

 

[qwk]

Less snippiness, please.

The BMS software is definitely a trade secret. How it tries to detect a chemical condition in some of the cells and the proper response to ensure safety and to ensure a longer life also are, since these details expose characteristics of the battery chemistry.

How was that snippy? Wow.

 

[qwk]

for safety reasons, to ensure less degradation of the battery in future...you name it, I have only guesses, but "Tesla does it just to spite their customers" is not one of them. Tesla isn't exposing this, but there must be some reason for capping Vmax and there must be some trigger.

Obviously, the packs that are affected are now "below average", but as said before, not everyone can have above average batteries.

Also, the BMS only has limited data so it's perfectly possible for some false positives to occur (i.e. for the BMS to cap a battery when in fact, if you knew exactly the state of the entire battery pack, that would be unnecessary).

And I'd expect them to first err on the side of caution and avoid false negatives (a BMS that would not cap Vmax when it would be the right thing to do) as a primary consideration, only tweaking to avoid false positives later.

Unfortunately devising tests that have no false positives and no false negatives is not possible.

Tesla could give two poops about safety of used cars. Now, warranty obligations, and paying for crispy critters is another matter...

 

[sixela]

That's all speculation and opinion and little fact...without even qualifiers to mark it as such.

 

[qwk]

For the last time, the exact analogy to this scenario would be this.

One goes to the store to buy widgets(I’ll use big screen TVs in this instance).

I go out and buy 265 big screens with an unlimited tv watching 8 year long warranty, also guaranteed in case of fire).

After 3-6 years, 15 no longer work, and you threw them out. You hear about Joe down the street have a big screen bought from the same store catch fire. Next thing you know, a truck pulls up in your driveway, and loads 40 of YOUR Tv’s up, and drives off. Remember they still were functional tv’s.

In anger, you call the store to find out what is going on, and they tell you it’s for your safety, and not to worry as you still have 210 tv’s left.

Who here thinks that this behavior is ok?

 

[DJRas]

Software limits don't equal degradation, but they can be a rational response of a BMS system to detected degradation.

I would say they detected a FAULT, not degradation.
The part of the battery that is NOW locked out was usable prior to May 15, 2019.