Wiki Sudden Loss Of Range With 2019.16.x Software

[JRP3]

So what you are saying is that if Tesla reinstated the original charging profile there would not be a problem that would warrant a recall?

That's essentially what wk057 is saying. Tesla lowered voltage until they could come up with a software fix to compensate for the faulty component. That's completely different than defective cells physically short circuiting.

 

[wk057]

No matter how you slice and dice the Batterygate saga, it was done for the safety reasons as most of us have believed form day one. You overcharge those cells you have fire.

Can a side effect of the employed "mitigation" be a correction of your displayed range? Sure, it can. But the primary reason was to stop the fire incidents, which we have not had since May 2019.

It was not. Nothing in pre or post 2019.16 firmware would allow overcharging cells, even if Condition X or Z were present. This is the point I was making. That portion of the firmware has not changed.

The "abundance of caution" update didn't appear until weeks later and is completely unrelated.

If you have any actual evidence to the contrary, I'm all ears. But you do not. I have a fully annotated RE of the BMS the says otherwise, along with some recently leaked internal documents that concur.... but sure, I'd love to see your evidence.

But you missed the main point: Tesla made that linkage in their statement following the fires.

As noted, Tesla's own versioning mechanism refutes this. 2019.16 is firmware built in the 16th week of the year, went through a QA process, and was released 3-4 weeks later. The "caution" update was kind of announced about the time 2019.16 was released, but that build didn't hit production cars until 3-4 weeks later.

And no one has yet explained the fires. So Tesla has still not unlinked them.

I can explain them completely, but they still have nothing to do with this thread.

 

[cousin_IT]

?? I didn't dislike your post ...

No you didn't, don't be offended. I'm just teasing some of the readers of this thread that disagree with every statement that's not in line with their opinion.

 

[hpartsch]

So what you are saying is that if Tesla reinstated the original charging profile there would not be a problem that would warrant a recall? Tesla has just changed the charging profile because they wanted to see how the owners would react or because an intern thought it would be fun?

But if no one outside Tesla knows what the underlying issue for the downgraded charging profile is, then how do you know is not the same issue, and not only not the same issue, but not 'the same issue at all'.

And even if it is not the same issue, it shows very different way of handling what ever issue it may be.

I like your thinking.

 

[Guy V]

They had a different and more severe issue than Tesla: "Hyundai said an investigation into the fires showed the cars' defective LG-made battery cells could short circuit."

No doubt they also expect significant cost recovery from LG. We'll see if they work this out between themselves or it becomes a major PR finger-pointing war. EVs aren't the total, core or even a critical part of Hyundai's business, but were it to happen, that kind of confrontation with Panasonic would be suicidal for Tesla. There is very scary potential now for a blowup at LG to cripple EV introduction for a huge part of the auto industry, and undermine EV confidence for everyone. This is something to watch closely.

 

[Guy V]

It was not. Nothing in pre or post 2019.16 firmware would allow overcharging cells, even if Condition X or Z were present. This is the point I was making. That portion of the firmware has not changed.

The "abundance of caution" update didn't appear until weeks later and is completely unrelated.

If you have any actual evidence to the contrary, I'm all ears. But you do not. I have a fully annotated RE of the BMS the says otherwise, along with some recently leaked internal documents that concur.... but sure, I'd love to see your evidence.



As noted, Tesla's own versioning mechanism refutes this. 2019.16 is firmware built in the 16th week of the year, went through a QA process, and was released 3-4 weeks later. The "caution" update was kind of announced about the time 2019.16 was released, but that build didn't hit production cars until 3-4 weeks later.



I can explain them completely, but they still have nothing to do with this thread.

WOW! Explaining the fires is an order of magnitude more significant than our battery gimping. Where and when are you going to?

Far, far more important, when is Tesla going to?

 

[MP3Mike]

How did Hyundai handle the same issue? Did they decide to basically steal from their customers or recall and replace 82,000 batteries? Well you can read here what they decided to do - and spoiler, Hyundai did not make the same decision as Tesla.

Hyundai's recall of 82,000 electric cars is one of the most expensive in history - CNN

The other interesting thing about this is that Hyundai tried a software fix first, but it didn't work and the fires continued. GM is having what appears to be a similar issue with the LG batteries in the Bolt EV, but they are still saying that they are going to fix the fire issue via software, hopefully in April. While their cells were made by LG, just like with the Hyundai ones that are defective, I think the ones GM used were made in a different plant so they might not have the same defect. (Of course that doesn't really fit, since late 2019 and newer Bolt EVs are not supposed to be impacted and the cells in them were made in yet a different LG plant in the US. So that makes it seem like it could be a defect in the cells, not the software.) Only time will tell if GM can safely fix their fire issue with a software update.

 

[wk057]

WOW! Explaining the fires is an order of magnitude more significant than our battery gimping. Where and when are you going to?

Far, far more important, when is Tesla going to?

At this rate, never. If folks can't even accept simple truths related to how and why the battery range issue was handled and how it's unrelated to any fire issue... then I'd expect even less of a reception to the truths of the battery fires... especially when I've managed to determine the exact issue, personally own a vehicle with the issue, and from that figured out exactly how many cars were potentially affected. Since that number is in the single digits (out of hundreds of thousands), and Tesla's already taken care of the issue with those vehicles... well, it seems highly unlikely that sharing those details would be of any benefit whatsoever. Considering those details will be even less verifiable by the masses than the details of my BMS RE efforts with regard to this range issue, and still so many refuse to accept it... I don't think it's worth getting involved.

And Tesla explaining? lol. Yeah, sure. Let's dig up a non-issue and bring media attention to it. Sounds like a WONDERFUL plan for them. /s

So I'm just going to leave it at my original note: Nothing whatsoever with this issue is in any way related to any risk of fire.

 

[DJRas]

I guess you didn't actually read Tesla's statement on May 15, 2019. It did not say they were starting to work on the problem. It did not say they will address the problem next month. It said
“As we continue our investigation of the root cause, out of an abundance of caution, we are revising charge and thermal management settings on Model S and Model X vehicles via an over-the-air software update that will begin rolling out today, to help further protect the battery and improve battery longevity.”

They certainly began looking into this weeks before May 15 to have something they could announce to be released that day. Certainly they had been working on it no later than April 22 when the Shanghai fire happened and more intensely after the SF fire May 5.

It was not. Nothing in pre or post 2019.16 firmware would allow overcharging cells, even if Condition X or Z were present. This is the point I was making. That portion of the firmware has not changed.

The "abundance of caution" update didn't appear until weeks later and is completely unrelated.

If you have any actual evidence to the contrary, I'm all ears. But you do not. I have a fully annotated RE of the BMS the says otherwise, along with some recently leaked internal documents that concur.... but sure, I'd love to see your evidence.



As noted, Tesla's own versioning mechanism refutes this. 2019.16 is firmware built in the 16th week of the year, went through a QA process, and was released 3-4 weeks later. The "caution" update was kind of announced about the time 2019.16 was released, but that build didn't hit production cars until 3-4 weeks later.



I can explain them completely, but they still have nothing to do with this thread.

 

[wk057]

I guess you didn't actually read Tesla's statement on May 15, 2019. It did not say they were starting to work on the problem. It did not say they will address the problem next month. It said
“As we continue our investigation of the root cause, out of an abundance of caution, we are revising charge and thermal management settings on Model S and Model X vehicles via an over-the-air software update that will begin rolling out today, to help further protect the battery and improve battery longevity.”

As noted in my PM:

Of course I read it. It just confirms the complete lack of internal communication that's persisted for the better part of the past decade with Tesla. The fact of the matter is that there is nothing in the 2019.16 that is "revising charge and thermal management settings". 2019.16 only adjusts the way the range algorithm gets its data, as well as how the BMS treats potentially erroneous BMB readings. Pre-2019.16 it wouldn't allow the effect of an anomalous reading to cause caps on the upper or lower end of the SoC range to stay in place once it received several hours worth of known-good data... post-2019.16 this timing was extended slightly to about a day in order to better search for vehicles with Condition X. None of this changed thermal settings, and none of the changes charge settings.

Thermal and charge settings were tweaked, however, in 2019.20... exactly in line with their normal release cadence.

I'll expand further: They explicitly state they're changing "thermal management settings"... yet none were changed in 2019.16... but both thermal management settings and charge settings were adjusted in 2019.20... exactly in line with when code built when the statement was made would be pushed to the public.

 

[Droschke]

The capping had nothing to do with the safety (fire prevention). It was for the displayed range correction which Tesla has already told us about it. Plain and simple. Not sure why the folks can't even accept simple truth /s

 

[wk057]

The capping had nothing to do with the safety (fire prevention). It was for the displayed range correction which Tesla has already told us about it. Plain and simple. Not sure why the folks can't even accept simple truth /s

Sarcasm aside, I'll gladly review whatever evidence you have that supports your narrative.

 

[Guy V]

At this rate, never. If folks can't even accept simple truths related to how and why the battery range issue was handled and how it's unrelated to any fire issue... then I'd expect even less of a reception to the truths of the battery fires... especially when I've managed to determine the exact issue, personally own a vehicle with the issue, and from that figured out exactly how many cars were potentially affected. Since that number is in the single digits (out of hundreds of thousands), and Tesla's already taken care of the issue with those vehicles... well, it seems highly unlikely that sharing those details would be of any benefit whatsoever. Considering those details will be even less verifiable by the masses than the details of my BMS RE efforts with regard to this range issue, and still so many refuse to accept it... I don't think it's worth getting involved.

And Tesla explaining? lol. Yeah, sure. Let's dig up a non-issue and bring media attention to it. Sounds like a WONDERFUL plan for them. /s

So I'm just going to leave it at my original note: Nothing whatsoever with this issue is in any way related to any risk of fire.

The only problem I had, and I suspect most here had, with accepting your simple truths is that you never gave them to us. Tesla still hasn't come forth with them.

I don't know about every one else here, but I would have been satisfied with a notification and promise of progress reports from Tesla that they had discovered instances of anomalous sensor readings in certain conditions in a small number of batteries and out of an abundance of caution were restricting maximum charging to guaranteed safe levels in those that might be affected while they worked to resolve the issue.

It's an axiom: You can't have trust in the decisions and pronouncements of people who hide the facts of the matter.

Another major aspect of this whole battery-gate saga is that Tesla still has not provided or given any hint of providing a practical-cost out-of-warranty battery replacement solution. Until they do we have to assume now that useful life of car = useful life of original battery. That will be how purchase decisions will have to be made. Once that becomes widely recognized it will inevitably become the new primary competitive measure. We think of batteries as disposables, and that will then carry over to comparisons of EVs versus ICE vehicles which we know to be all too endlessly repairable.

 

[Chaserr]

But you missed the main point: Tesla made that linkage in their statement following the fires.

And no one has yet explained the fires. So Tesla has still not unlinked them.

we know this has always been about fire safety. If it wasn't, Tesla is in big trouble for not disclosing their other fire investigation and they still don't have any kind of mitigation to stop the danger!

 

[lightningltd]

Can we get a facepalm button in addition to the like / dislike ones?

 

[JRP3]

Until they do we have to assume now that useful life of car = useful life of original battery.

That seems like a stretch. You can always get a low mileage battery from a salvage vehicle even if Tesla won't sell you a new one.

 

[wk057]

The only problem I had, and I suspect most here had, with accepting your simple truths is that you never gave them to us. Tesla still hasn't come forth with them.

I don't know about every one else here, but I would have been satisfied with a notification and promise of progress reports from Tesla that they had discovered instances of anomalous sensor readings in certain conditions in a small number of batteries and out of an abundance of caution were restricting maximum charging to guaranteed safe levels in those that might be affected while they worked to resolve the issue.

It's an axiom: You can't have trust in the decisions and pronouncements of people who hide the facts of the matter.

Another major aspect of this whole battery-gate saga is that Tesla still has not provided or given any hint of providing a practical-cost out-of-warranty battery replacement solution. Until they do we have to assume now that useful life of car = useful life of original battery. That will be how purchase decisions will have to be made. Once that becomes widely recognized it will inevitably become the new primary competitive measure. We think of batteries as disposables, and that will then carry over to comparisons of EVs versus ICE vehicles which we know to be all too endlessly repairable.

Fortunately I have a few "axioms" on my side here:

Facts are true regardless of if you believe them or not. Facts don't care about your feelings. Facts don't need your approval to be true.

And that's where I'm at with all of this. It doesn't matter what anyone here believes or otherwise refuses to accept at this point. The fact of the matter is that this has nothing to do with fires. Nothing whatsoever. Doesn't matter if Tesla says anything about it or not, it's still true. This is verified by a third party (me), and could certainly be re-verified by other third parties with the skills required to put in the reverse engineering effort. That's science. Apply the scientific method and first principles to the problem, and come up with a reasonable and verifiable conclusion.

Thus far no one has presented anything whatsoever that refutes the conclusions drawn in my writeup. In fact, everything I've seen since has supported those findings. (No, just disagreeing or complaining doesn't count as evidence.)

No amount of jumping up and down handwaving is going to make this related to fires. It's time people get over that.

 

[Droschke]

There is no secrete that the matter is under litigation and the facts are not public yet. Someone's opinion on Internet is not those facts.

Edit:
------
I appreciate the efforts spent in the analysis, but I respectfully and strongly disagree with the conclusion reached as "fact".

 

[Battpower]

I know this has been said before (and I'd be happy to face palm myself) but why would Tesla be abundantly cautious about no elevated risk? (regardless of exact timing and release of mitigation / diagnostic measures).

And having concluded their protracted analysis of the non-risk and come up with a work around that could take months to reach calibration - more kicking the can down the road - (if I understood that correctly) why would they not make a clear statement to those owners impacted along with a warranty extension or other similar gesture to back up their words? [edit: at the very least the warranty should be extended until the sticking plaster has demonstrably returned performance back to pre-capped level.]

And while they are wrapping up the messaging over that one, why not address the reduced charge rates with some factual data that explains exactly why the charge rate reduction is applied and why it was necessary? ie what would the implications be if charge rate was left as per original spec?

What did the owners of those effected cars do wrong that resulted in their car’s performance being restricted?

Edit: And, if no mitigation measures had been put in place on a particular car, could that car be less safe than a mitigated car? (I would say potential sudden unexpected shutdown would be a safety issue for example). Have there been any safety / mandatory software updates issued? If not, that wouldn't seem abundantly cautious until you had your diagnostic data back.

 

[Droschke]

I know this has been said before (and I'd be happy to face palm myself) but why would Tesla be abundantly cautious about no elevated risk? (regardless of exact timing and release of mitigation / diagnostic measures).

And having concluded their protracted analysis of the non-risk and come up with a work around that could take months to reach calibration - more kicking the can down the road - (if I understood that correctly) why would they not make a clear statement to those owners impacted along with a warranty extension or other similar gesture to back up their words?

And while they are wrapping up the messaging over that one, why not address the reduced charge rates with some factual data that explains exactly why the charge rate reduction is applied and why it was necessary?

What did the owners of those effected cars do wrong that resulted in their car’s performance being restricted?

Edit: And, if no mitigation measures had been put in place on a particular car, could that car be less safe than a mitigated car? (I would say potential sudden unexpected shutdown would be a safety issue for example). Have there been any safety / mandatory software updates issued? If not, that wouldn't seem abundantly cautious until you had your diagnostic data back.

These are all excellent points. Thanks for not being easily persuaded to believe someone's thesis as facts.