Wiki Sudden Loss Of Range With 2019.16.x Software

[Droschke]

you don't make up a bunch of nonsense, post that as if it were true with zero qualifiers, and expect someone with knowledge to correct it for you

Instead, you do what normal people do: ask.

Fine. Asking ... what are conditions X and Z?

Speculations:
- Li-Plating
- Dendrites
- Dead cells

You claim to have the knowledge. Correct it for us, please.

 

[johnman10]

This is all very cute, but the truth of the matter is that Tesla has treated us early owners like livestock and is pretending that everything is fine. There is no debate here. It is us, trying to protect ourselves from what turned out to be a shitty corporation. I don't care if it is dendrites, I don't care if it is the cooling system, I don't care if it is some aliens sending EMP to our cars trying to frame Elon. All I care about is that I trusted lying crooks who try really hard to benefit at the expense of my bank account.

 

[Guy V]

Wait wait... now the excuse for the misinformation is that people are posting misinformation in order to provoke others into correcting that information? lol

Like, for real... if you want to know something technical about how something works that you don't understand.... you don't make up a bunch of nonsense, post that as if it were true with zero qualifiers, and expect someone with knowledge to correct it for you (All while your misinformation is read and accepted by others who don't know any better). Instead, you do what normal people do: ask.

Again, speculation posted as fact is inexcusable. There's no references to excuse any of the things I've quoted in my other post. These things are either blatant lies, or otherwise require some citation. It isn't much to ask that a) people don't lie, and b) people cite sources for things they can't possibly have first hand control of (like, you know, claiming there's a preposterous Federal law about something... maybe post a link to the law if it actually exists?)

But that IS speculation, stringing together bits of information gleaned from any available sources and offering it for comment and constructive criticism. That's exactly how the scientific method and peer review works.

We all understand that Federal testing must include required defined conditions for it to be uniform, measurable and meaningful. It seems you are making the preposterous claim that they don't. If you do not believe the charged voltage levels of the battery is part of that, then YOU dig out the full criteria and refute that. Changing the normal operating parameters in software from those used in EPA testing is precisely what has cost VW $33 Billion.

 

[Guy V]

Can someone please PM me when all the arguing and poo flinging is done? Getting tired of cleaning all the crap off my screen
View attachment 570628

No worries, one way or the other it will most certainly follow some big announcement from Tesla.

 

[Battpower]

Speculations:
- Li-Plating
- Dendrites
- Dead cells

- Sub-par cell manufacture.
- Uneven cell degradation.
- Material / chemical issue (at manufacture)
- Under-spec'd cooling system.
- Poor cell matching.
- Short term cell supply chain issue.
- Poor design / under spec'd components.

I think all of these have been touched on with some justification.

 

[Battpower]

the thermal control system allowed it to gradually increase the pack temp naturally while charging, and eventually the taper in current meant less heating in general to deal with anyway, and thus less cooling needed.

Given the situations I observed and the anicdotal nature / no hard facts, this is about as near as anyone is likely to get and quite possibly fits reasonably well.

I would add though that it did seem very odd / noticeable that all bar an odd one or two of the cars I observed appeared to be cooling aggressively. The SOC's could have been different, or I could have been charging slower due to not pre heating or not pushing the car as hard.

However, all cars in France drive at 130kmph / 80mph on autoroute. It's why you pay so much in tolls. So my battery not cool, but didn't have SMT at that time.

There's a ton of factors in determining how much thermal control is needed.

Yes. Although it might seem like a waste, would you list as many as you can think of? (No description but as little overlap as possible).

 

[David99]

Error message from an affected battery pack. This should clear up some of the speculation. The key word 'hidden' means these error messages only show up when the car is in service mode.

In my opinion...
#2 it clearly states the voltage limitation.
#1 'weak short' appears to be dendrites (dendrites are needle-like and can poke through the insulation layer).

I would highly recommend updating the software on your car so they can better diagnose your battery and apply the appropriate amount of protection or possibly even remove the limitations.

 

[Battpower]

Cells develop more internal resistance as they age, and this resistance creates heat which must be dissipated.

In general terms I also believe this to be true. Obviously we are speculating about something that is seen by some as 'my range has dropped', others as 'my car could burst into flames any minute, or my battery doesn't charge to the same voltage as it used to or it takes all day to charge etc..... Behind all these is a technical reason / reasons that may be unavoidable or not. Inter-related or coincidental.

By Tesla they are seen as something they need to find a way to explain and own up to without screwing themselves too badly...

Then there is every single twist, nuance, quirk, design element, component that makes up the manuci of the battery.

So we need to be detailed, accurate and specific about something we don't (and to some extent can't / aren't allowed) to understand.

So, is it true to say that all lithium cells degrade with age, use patterns and net energy throughput?

Is it true that lithium cells have multiple aging 'modes'?

If so, what are these modes?

Maybe:
Towards lower internal resistance -
Plating
Dendrites
? Thermal related behavior?

Towards higher resistance -
(I don't know what these effects are called or even if the effects exist)
Electrolyte aging.
Electrode decay / chemical change.
Age related increase in internal resistance.
Temperature accelerated degradation.

Are there any other modes?
Eg: mechanical? Is Dendrites electrical or mechanical?

Functional? 'operating stress' required in battery to maintain acceptable performance. Like a redundant system is only redundant until the redundancy is not available. So the battery is able to remain functional (safely) while it has not experienced degradation.

Is it possible / likely that cells (used by Tesla) exhibit a mix of 'aging' effects towards both ends of the spectrum?

Have bricks (blocks of parallel cells) ideally from a battery showing signs of stress) been inspected with public results showing any data like:

Number of blown cell fuses.
Condition of cells remaining connected and (presumably) sharing same charge / discharge current as a good cell but between (far?) fewer cells?

Is this the kind of situation (bad bricks) that the 'diagnostic' software Tesla rolled out was looking for?

If not, then what other battery behavior could Tesla usefully have looked for to tell them something they didn't already know about battery condition?

 

[Droschke]

Error message from an affected battery pack. This should clear up some of the speculation. The key word 'hidden' means these error messages only show up when the car is in service mode.

In my opinion...
#2 it clearly states the voltage limitation.
#1 'weak short' appears to be dendrites (dendrites are needle-like and can poke through the insulation layer).View attachment 570684

I would highly recommend updating the software on your car so they can better diagnose your battery and apply the appropriate amount of protection or possibly even remove the limitations.

Thanks David. Do you know if Tesla is replacing that pack?

 

[David99]

Thanks David. Do you know if Tesla is replacing that pack?

So far they have not offered me a replacement pack.

 

[gmo43]

What? How do you dismiss these factually false statements as "speculative hypotheses"? Like, they're not even speculation... they're things that are obviously wrong on a factual and technical level.

I would _love_ to see this thread get back on track, but that's basically impossible with people posting so much nonsense.

If you would like to see this thread get back on track then maybe guide us in the right direction since you don't want to tell us what you know. Since all we got is speculation on our side.

 

[Guy V]

Error message from an affected battery pack. This should clear up some of the speculation. The key word 'hidden' means these error messages only show up when the car is in service mode.

In my opinion...
#2 it clearly states the voltage limitation.
#1 'weak short' appears to be dendrites (dendrites are needle-like and can poke through the insulation layer).View attachment 570684

I would highly recommend updating the software on your car so they can better diagnose your battery and apply the appropriate amount of protection or possibly even remove the limitations.

That's how I would interpret it.

Perhaps someone could explain how that matches up with this claim:

"This is false. This is not at all how software locked packs performed capacity capping. This is misinformation represented as fact. (More detail: The BMS capped capacity, not voltage, and scaled that locked cap with measured degradation. Charge end voltage was not capped in any way, and in fact would usually change from charge to charge.)"

 

[Droschke]

So far they have not offered me a replacement pack.

Thanks. I did not know that was your pack. Hope they replace it for you. I'm sure you will keep us updated.

 

[DILIE]

Yes please!

Any suggestions where to look for some more worthwhile / fertile ground to discuss / investigate? Are there any of the recent posts that strike you as 'interesting' / relevent?

Not new, but the apparent 'obsession' with keeping certain packs as cool as possible under certain charge / temperature conditions is not just chance. So what else can we link to that?

Early this year I did a long trip through France. The weather was not hot, and I noticed that pretty well all the MS & MX's being supercharged on the Autoroutes had the cooling system fans running flat out, compared with my MS R LR that charged in near silence from 20 to 80+ percent.

Is it all a response to normal degradation? Forget for a moment to issue of OTA updates materially changing a car's performance. Just purely from a technical view point, what are the most likely candidates / explanations as to why my pack charged with apparently little cooling while many others apparently justified so much?

Am I confused? Your Raven has a 100kWh battery, right? Those are not known to be affected by charge limiting or extra battery cooling, or yours would be the first known report. The other cars are affected by the 2019.16 or later changes for 85kWh and smaller batteries. The fan behavior is not strictly tied to battery capping. My car roars like a Concorde taking flight, but is still, as yet, uncapped.
I remember the first trip down I-45 after the change, walking out of The Collin Street Bakery in Corsicana after lunch and noticing other people staring at the car as they walked by. Also drove to a Cowboys game and enjoyed the curious stares as I navigated amongst the pedestrians.
I hadn't realized it was tied to charge, until someone mentioned it here and I realized, "Oh yeah, I haven't heard the roar coming from the garage since I lowered the charge limit to 70%.

 

[qwk]

That's how I would interpret it.

Perhaps someone could explain how that matches up with this claim:

"This is false. This is not at all how software locked packs performed capacity capping. This is misinformation represented as fact. (More detail: The BMS capped capacity, not voltage, and scaled that locked cap with measured degradation. Charge end voltage was not capped in any way, and in fact would usually change from charge to charge.)"

Jason is being intentionally vague IMHO. Tesla applied the cap by limiting capacity, but they could have easily capped the voltage instead. Not much difference. The end result is similar. Voltage is a big part of the rated range algorithm, so I’m not sure why he insists that it doesn’t matter that capped cars cannot charge to 4.2 V.
The issue here is that the software was manipulated by Tesla, which in turn stole what was paid for initially. This was done solely for Tesla’s benefit, without the owners permission.

BTW, the 60-40kwh battery capacity cap On the early 40kwh cars was done the exact same way.

 

[Battpower]

explanations as to why my pack charged with apparently little cooling while many others apparently justified so much?

yours would be the first known report.

To clarify, I was just drawing attention to an obvious difference between my newer car and a random selection of other (generally older) cars being supercharged.

My car roars like a Concorde taking flight,

Describes perfectly how several cars at multiple superchargers all sounded while mine charged in silence. Given the way that most cars are driven on French autoroutes and the long distances between junctions, it was quite likely that most would have been cruising at 80mph for at least 15 miles immediately prior to charging. Of course there is no reliable evidence as to the SOC of the observed cars, but again the fact that all vehicles in question were paying tolls to expedite their journeys would probably go against sitting (needlessly) to charge (slowly) at an already high SOC.

vague IMHO. Tesla applied the cap by limiting capacity, but they could have easily capped the voltage instead.

From: Battery Voltage Information – Battery University

"Battery users want to know if Li-ion cells with higher charge voltages compromise longevity and safety. There is limited information available but what is known is that, yes, these batteries have a shorter cycle life than a regular Li-ion; the calendar life can also be less. Since these batteries are mostly used in consumer products, the longevity can be harmonized with obsolescence, making a shorter battery life acceptable. The benefit is longer a runtime...... "

While BMS's can be super complex, in essence they are simple. Maintain a battery of cells within safe operating limits while optimising energy storage.

The question of capping charge vs voltage is imo moot since charge is a flow of electrons (current) for a period of time and you need a difference in 'pressure' (voltage) to make the current flow. (The SI system defines the coulomb in terms of the ampere and second: 1 C = 1 A × 1 s). I would argue that you cap the voltage to protect the cells by stopping the flow of current. As you get closer to the maximum permitted cell voltage you have to reduce the current (slowing down the charge) but afaik the ultimate factor that must be monitored to maintain safety is the voltage. It is likely that by the time the charge rate has been reduced to a very low level, the 'diminishing return on investment' (of time and net energy input) make it pointless continuing to charge even if there were no potential negative side effects.

 

[Battpower]

BMS doesn't use voltage at all for capacity measurement and instead uses coulomb counting

Bold added by me.

Is 'coulomb counting' more than measuring current and time?

This would measure charge flow and is effectively a 'capacity' measurement.

But capacity measurement is not the same as protecting cells from going above a set voltage - regardless of charge factors.

What does the Tesla BMS actually do? Based on readings from SMT it tries to get cell / brick voltages balanced at a set level. Best I have seen is 3mV and worst around 15 mV with pack idle and minimal current through pack.

It seems to balance by bringing all cells to the same voltage rather than the same amount of charge flowing into each cell. In fact, with the parallel cells in bricks, I would argue that there is no mechanism to control or measure charge flow in each individual cell. That will be determined by the characteristics / state of each cell.

It seems to me that while you can use the voltages (? and in-use temperature readings?) of series connected bricks to guess at / predict the state of each brick, you have no such means of comparing the condition of individual cells within bricks since they are forced to the same voltage by being in parallel.

Since every single cell is wired in a brick, ISTR that other than by hoping / assuming / wanting / manufacturing cells to be as near as possible identical both on day 1 and towards EOL, you actually have no way of KNOWING how the various levels of degradation are effecting charge & load sharing between cells.

Specifically:

- Is there any way (in normal operation) that Tesla can measure or control charge flow / energy balance between individual cells?

- Is it true or false that cells can and probably do degrade at different rates (and possibly in different ways between cells) that could / would exasebate subsequent imbalance and degradation)?

@wk057 (or anyone else suitably set up), could you use a thermal imaging camera to view a heat map of parallel cells in bricks being charged / under load?

This would possibly give some insight at to conditions within typical bricks.

 

[Battpower]

That's how I would interpret it.

Perhaps someone could explain how that matches up with this claim:

"This is false. This is not at all how software locked packs performed capacity capping. This is misinformation represented as fact. (More detail: The BMS capped capacity, not voltage, and scaled that locked cap with measured degradation. Charge end voltage was not capped in any way, and in fact would usually change from charge to charge.)"

Likewise, I struggled to understand that because of the apparent contradiction.

 

[raphy3]

Not challenging this ^^^ at all, and accepting the non-liniarity of charge vs volts, which (if either? or both?) is more responsible for stress on the battery? Charge or voltage?

And if it's not voltage, and Tesla BMS actually counts charge, why cap the voltage rather than limit charge?

With all other things being constant, current and voltage are inextricably linked. It really isn't as irrelevant as wk057 seems to be trying to make it sound to us.

To raise the voltage of a battery terminal, actual charge must be stored there. Higher voltage relative to the other terminal means that terminal is more positively charged than the other. Storing lots of charge for long periods (or even short periods) leads to cracking of the anode as the volume of the anode physically changes. Not saying that Tesla reduced voltage to reduce anode fracturing, but if you reduce the voltage (the potential difference between the cathode and anode) of a li-ion battery, you will significantly increase the number of charge/discharge cycles the battery can handle.

 

[wk057]

Read through the last few pages. Less noise surprisingly. Still no one has it right, though. *shrugs*

Old pic, but 5x 85-type modules being charged at 1/2C (~40kW equiv) for about 30 minutes. They heat evenly even with no liquid cooling.

Did anyone know there are something like ~400 MOSFETs inside the battery pack, not even counting the main BMS? Just a fun note.