Wiki Sudden Loss Of Range With 2019.16.x Software

[Chaserr]

I wonder what makes say the 100 packs any better? Is it better thermal management keeps pack temps lower, thus aiding the balancing dissipation? Or better matched cells that don't need as much balancing? Or just putting more stringent control on operating loads so the cells gets less stress?

100 packs have completely different hardware designs and probably a smarter setup managing safe balancing of cells, temperature, and anything else they learned since the first S batteries were installed.

They are also electrochemically different internally and chemical makeup is now known to be extremely important in mitigating lithium plating, even at higher C rates and temperatures than Tesla's original chemistry.
https://aip.scitation.org/doi/full/10.1063/5.0015099
Additives may be present in the newer cells to improve safety and inhibit lithium's natural tendencies to form dangerous SEI layers under high energy conditions.

We will learn if these problems were fixed if the newest Long Range packs and Model 3s stop charging consistently at their peak advertised kW rates or remain useful longer than the 85.

 

[Battpower]

But, that theory doesn't seem to make too much sense because the voltage caps being applied are only like 3%.

Remember we are potentially looking at multiple issues. The voltage cap is imo so far most likely related to internal cell chemisty although of course could have broader relevance. In an extreme case, your weakest brick could hit Vcap long before any other bricks, so the additional balancing dissipation effect would apply to only one brick. I haven't really thought that scenario through, but don't think Vcap would necessarily be involved in dissipation demands.

Detecting electrical shorts and hiding them so they aren't repaired is a safety violation. Intentionally reducing performance to save money on warranty repairs is a warranty violation.

I can see the warranty argument but at the moment not quite so clear safety. Would need to run a battery with plating and whatever other changes in composition to be run at high voltages to see the outcome.

Do we know that there is no improved cooling on the module boards in the new packs?

That also was one of the points I want to find out. So far I haven't found pictures of 100 packs. As long as bricks stay sufficiently matched, maybe there is no problem. Maybe Tesla think they now better understand how they can manage this.

Edit: per @Chaserr post, several changes have been made.

 

[Battpower]

We will learn if these problems were fixed if the newest Long Range packs and Model 3s stop charging consistently at their peak advertised kW rates

Did you look at the TeslaLogger charts for other cars a few pages back? The impression I got was that the trend is to allow peak charge power for quite narrow SOC ranges. Again I suspect there are too many factors behind those charts to interpret them with certainty, but certainly newer cars did not appear to me to have extended periods of high charge rates. Obviously Tesla is trying to find the optimal limits they can push too.

 

[Chaserr]

Are charge times being slowed already on these new cars? That isn't what I was hoping to learn, but we are learning from these answers sooner than I expected.

Would need to run a battery with plating and whatever other changes in composition to be run at high voltages to see the outcome.

Identification of Lithium Plating in Lithium-Ion Batteries by Electrical and Optical Methods - IOPscience

This became a well documented field of study in the year since Batterygate began. Plating can be mostly managed and reversed if it was accounted for in a new battery, but progression becomes impossible to manage eventually, especially in cells that hadn't been managing SEI buildup for most of the last 8 years or so.

Chargegate and draingate might be about trying to manage SEI on older cells, reducing the potential for fires yet again. I still hope not because we haven't been informed which is still a crime all by itself, but I hold out hope they have a plan to eventually come clean and beg forgiveness. Hardware problems from unmanaged SEI are why we bought the warranty. Mitigating warranty costs is a lesser crime (in my opinion, I don't know if federal law sees it that way but I hope so) but they still need to fix what they broke.

 

[David99]

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

There are a few people that know exactly what is going on, but due to the ongoing lawsuit they are not talking/posting.

I have nothing but respect for your knowledge and pioneering work you have done around Tesla cars. I appreciate everything you have shared in the last years! But poking in here every once in a while teasing everyone that you know but wont say anything is a little childish.

I was the first person to help David R. read the data from his car, helped him set up his own connection. I shared years worth of CAN bus data collected by me with him to help make his case. I shared the error message about the 'weak short' using root access to my car. I was debating if I should share it or not because I feared Tesla would flag my car for whatever they want to do to punish me. But I put it out anyway. Don't get me wrong, you coming in here poking fun at everyone how they have no clue and bragging how much you know is a little off putting.

 

[JRP3]

Is it just that there is too much labor involved in bringing in thousands of cars to disassemble their packs and replace these boards?

If there are 400 or so boards I'd imagine so.

 

[Battpower]

how they have no clue and bragging how much you know

It is indeed off-putting. Trying to work this stuff out 'in public' with so little hard evidence and based on sometimes contradictory and misleading hints / clues is not easy. Personally I would hate to mislead or misrepresent, but I also want to know what is going on.

I mean we must have covered most bases somewhere in the thread. It's hard to imagine what difference a bit of certainty would make.

 

[Droschke]

There are a few people that know exactly what is going on, but due to the ongoing lawsuit they are not talking/posting.

I have nothing but respect for your knowledge and pioneering work you have done around Tesla cars. I appreciate everything you have shared in the last years! But poking in here every once in a while teasing everyone that you know but wont say anything is a little childish.

I was the first person to help David R. read the data from his car, helped him set up his own connection. I shared years worth of CAN bus data collected by me with him to help make his case. I shared the error message about the 'weak short' using root access to my car. I was debating if I should share it or not because I feared Tesla would flag my car for whatever they want to do to punish me. But I put it out anyway. Don't get me wrong, you coming in here poking fun at everyone how they have no clue and bragging how much you know is a little off putting.

Thanks so much, David, for stating what's on the minds of so many owners who follow this thread closely. I'm sure the affected owners appreciate your insightful contribution to this thread.

 

[bhzmark]

There are a few people that know exactly what is going on, but due to the ongoing lawsuit they are not talking/posting.

I have nothing but respect for your knowledge and pioneering work you have done around Tesla cars. I appreciate everything you have shared in the last years! But poking in here every once in a while teasing everyone that you know but wont say anything is a little childish.

I was the first person to help David R. read the data from his car, helped him set up his own connection. I shared years worth of CAN bus data collected by me with him to help make his case. I shared the error message about the 'weak short' using root access to my car. I was debating if I should share it or not because I feared Tesla would flag my car for whatever they want to do to punish me. But I put it out anyway. Don't get me wrong, you coming in here poking fun at everyone how they have no clue and bragging how much you know is a little off putting.

Being put off a little is a small price to pay for the information rec'd in exchange. I'd happily pay the price -- except I'm not put off at all.

 

[IngTH]

The links for fires posted by ingTH are all accident related and totally irrelevant.

Here are the links i was referring to:
Guangzhou: 广州一辆Model S发生自燃，没有充电，也没有发生碰撞-锂电池-电池中国网
Shanghai (well known): Tesla spontaneously combusts in Shanghai parking garage
Hong Kong: 【獨家】Tesla停車30分鐘後離奇自焚爆炸 車主逃過一劫

I have >20 fires while parking in my list.

A Nio with NCM811 caught fire around that time too, so the chinese authorities started investigation. Maybe the chinese GB/T 38031 safety norm which will get effective 2021 is a result of this investigation too?

 

[Battpower]

Are charge times being slowed already on these new cars? That isn't what I was hoping to learn, but we are learning from these answers sooner than I expected.

Random example:

Interestingly I have a 2020 X with the "F battery but used a 250kW charger for the first time and actually saw 200kW peak charging. That would be great if it hung around a little longer and did not drop so fast.

 

[Battpower]

Is it just that there is too much labor involved in bringing in thousands of cars to disassemble their packs and replace these boards?

We are possibly down a rabbit hole, so probably not worth much further speculation, but IF balancing components fail or become less effective (say due to elevated temperatures) then it may require bricks / modules to be replaced concurrently with balancing components, especially if the aging bricks / cells contributed to balancing component issue.

Given the importance of having closely matched bricks for optimal energy storage, it must be difficult to slot used or new modules into an old battery without potentially putting high demands on balancing. I expect the best source of replacement bricks would be old cars, since you need cells using same chemistry and probably having had roughly matching useage. Supply of suitable replacement cells to remanufacture batteries could be a problem.

I think these would be the likely contributing reasons for Tesla not jumping at wholesale component replacement.

 

[Chaserr]

Maybe the chinese GB/T 38031 safety norm which will get effective 2021 is a result of this investigation

GB 38031-2020. Auto-delivery. Electric vehicles traction battery safety requirements

 

[Spacey73]

I can't believe that an 82kwh pack is balanced by 100ma. I've just built a 1.8kwh powerwall for my solar panels out of 18650 cells and went with a capaciter based balancing system that takes the highest voltage of a cell pack and syphons off up to 5,000 ma to the lowest voltage cell pack.

It balances so fast and keeps uneven capacity packs in line within 0.01V even under load.

 

[JRP3]

On the other hand, i can say with high confidence that virtually all battery fires started in stationary batteries, regardless of their velocity in the preceding seconds.

The Mary McCormak car and I believe at least one other were being driven when they caught fire, and not in an accident.

 

[Battpower]

I can't believe that an 82kwh pack is balanced by 100ma.

You'd think with the load transients in an EV it would be even more important to be able to apply transient balancing to match.

 

[Spacey73]

True, I mean... I've built about a hundred 18650 Lithium Ion battery packs from tiny 250w output ones to 20kw beast output from a tiny 3kwh battery pack and the killer is when they become too low around the 3.2v and below packs start to get seriously out of balance.

The resistor based 0.25w per series ( so 14 x 0.25 w in a pack of balancing power) was nowhere near good enough and I would have to every couple of months individually charge each series pack to 4.1v to assure a relative balance.

You are only as strong as your weakest cell. There is no way a big pack in the Model S is balanced by a lowly 100ma, and these cars are so rarely at 4.2v which is where most balancers do their work....unless Tesla is constantly balancing at any voltage. You would have to leave your car at 100% for a month to even make a tiny bit of difference.

 

[Battpower]

unless Tesla is constantly balancing at any voltage

I'm not sure about 'any voltage', but certainly balances from 50% SOC. Balancing duration, predicting future need accurately ahead of balancing and incredibly well matched cells seem to me to be their salvation.

It may be that Teslas approach was based around cell quality being the holy grail for various reasons, then design a balancing system to suit. Obviously I don't know the design constraints, but dissipating imbalance rather than shuttling it from one brick to another seems short sited.

Maybe there is a job for Maxwell?

 

[islandbayy]

True, I mean... I've built about a hundred 18650 Lithium Ion battery packs from tiny 250w output ones to 20kw beast output from a tiny 3kwh battery pack and the killer is when they become too low around the 3.2v and below packs start to get seriously out of balance.

The resistor based 0.25w per series ( so 14 x 0.25 w in a pack of balancing power) was nowhere near good enough and I would have to every couple of months individually charge each series pack to 4.1v to assure a relative balance.

You are only as strong as your weakest cell. There is no way a big pack in the Model S is balanced by a lowly 100ma, and these cars are so rarely at 4.2v which is where most balancers do their work....unless Tesla is constantly balancing at any voltage. You would have to leave your car at 100% for a month to even make a tiny bit of difference.

They used to only balance above say, 90% SOC, but now they are balancing constantly while charging, at any SOC.

 

[Spacey73]

You say that they balance from 50% If this is true then that would help an awful lot in balancing a big pack.