Wiki Sudden Loss Of Range With 2019.16.x Software

[neroden]

NO change. They did not even bother to fix the safety issue I spent 45 minutes on the phone reporting from 2019.24.1
No surprise there.

OT, but could you start a thread on that safety issue?

 

[neroden]

I have a arbitration hearing date of 7 Aug. I also have an initial response from tesla managing counsel, product liability at tesla, maintaining all is well and that range calculation is complex and that I have a “potential misunderstanding “.

Does anyone have a copy of the battery warranty that was in place in November 2014?

Tesla counsel’s exhibit of the warranty provided in their response includes the Model 3.

Feb 2013 Warranty attached. Dunno if they changed it between then and Nov 2014, but you can definitely point out that Tesla counsel's exhibit is less likely to be accurate than this one.

Unfortunately, they literally exclude all damage caused by "usage of the battery".

 

[neroden]

Exactly but it seems like they're trying to make a convoluted argument to obfuscate or hide the fact that they are simply not charging to existing capacity in order to avoid having to argue WHY they aren't charging to existing capacity. If not fully charging point is stipulated, then they have to disclose why they software limited capacity. Did they do it to promote longevity? Did they do it for safety?

If they answer longevity, which is what they've told customers, then they're acknowledging that they have restricted full charge access to prevent FUTURE degradation and that the reduction is NOT really current degradation.

If they answer safety then they have to explain that they're limiting charging in order to prevent battery fires (aka Note 7) which is still not normal degradation and they have to admit that the batteries might catch fire even if they are less likely to do so on a lower SOC. If I were them I'd not want to even start down that road.

Of course their lawyers are ABSOLUTELY monitoring this thread which means their legal roadmap will influenced by what's discussed here which gives them a huge advantage.

Tesla's lawyers are flat out incompetent.

They've lost cases they really should have won in the past, like the trade libel case in the UK. They spent several years losing the case over Tesla's copryright infringements of GPL-licensed software, which they should never have tried to fight (it costs next to nothing to comply with the license). They are losing arbitration cases over trying to deny people replacements of deliminating touchscreens (I don't even know why they tried to fight those cases, they had no chance).

I don't think they're monitoring this thread effectively, even if they're monitoring it. I mean, for goodness sake, they filed the WRONG version of the warranty in one of the cases we're talking about, already. That's just an incompetent lawyer.

 

[egn1]

The screen shots we're posting are of static voltage at each SOC, NOT while charging.

Static voltage is irrelevant when charging and stopping charging at a specific voltage.

 

[jensk2]

I don’t know exactly what Vmin corresponds to but I have a TM-Spy screenshot from when my battery was at 0% and I coasted into a Supercharger parking lot.
View attachment 433393

According to this YouTube 'test' it is exactly 3.0 V Average (Matches your experience, you have 388,57 V which is 3.006V per cell:

 

[ytwytw]

What year is your car and current mileage ?

2017 100D
70000 km so far

 

[ytwytw]

What's your cell voltage at 100% or 90% as displayed on your IC? i.e. I ask that so we can see that this is software limited capacity rather than actual degradation.

Great Question.

Let me try to get those this weekend

 

[DJRas]

I agree that the way to fight this is the artificial battery capacity limit put on our affected batteries.

The fact that Tesla has sold cars with stated 60kWh capacity but having 70kWh packs at the same price as cars with actual 60kWh capacity batteries. And then later for a fee ($5,000?) was able to unlimit that capacity (over the air?) demonstrates the VALUE loss to those of us affected.

I have data from my car charging to 90% and to 100% that shows I can never get to the 4.2 volt 100% charge.

I am offering to either personally visit any affected owner in Southern California or send my CANBUS reader and a cheap smartphone to anyone else to capture similar data from as many cars as I can.

 

[MP3Mike]

NO change. They did not even bother to fix the safety issue I spent 45 minutes on the phone reporting from 2019.24.1
No surprise there.

@lightningltd Don't you think you should share the safety issue you found with people here so that they are aware and can deal with it? (Possibly getting more people to report it to speed up Tesla's response.)

 

[mjmiron]

2017 100D
70000 km so far

what was your car charging to prior to the update and after?

 

[faughtz]

If it was me proving my case I would stick to using voltage and kwh data whenever possible. It's too easy for tesla to make "range" sound subjective

That’s my approach.

 

[faughtz]

Feb 2013 Warranty attached. Dunno if they changed it between then and Nov 2014, but you can definitely point out that Tesla counsel's exhibit is less likely to be accurate than this one.

Unfortunately, they literally exclude all damage caused by "usage of the battery".

Thanks- I’m astonished that they would attach the current warranty.

The entire response was not responsive to my complaint, and has many inaccuracies.

 

[DJRas]

Since I don't yet have the offending change I still have all of my range(minus my 11 miles of "normal" "natural" degradation). I'm assuming that I'm going to be significantly impacted when I eventually have v9 forced on me for whatever reason so I'm getting all my data ducks in a row from Teslafi and CANBUS scans.

I just got on the waiting list for a C8. Going to do Museum Delivery this time in Bowling Green.

So, you say you are 11 miles degraded.
Does that mean your full rated range is 254 (down from 265)?
If so, then you are actually degraded by more because of their change to 276 Wh/mi.
The 265 was based on 295 Wh/mi (78.1 kWh usable)
Your displayed 254 is at 276 Wh/mi (70.1 kWh usable)
So 70.1 kWh / 295 Wh/mi = 237 miles using the same rating method.
265-238 = 27 miles actual degradation (10%)

THIS is why we need to talk about capacity in kWh AND max charge voltage.

 

[Ferrycraigs]

THIS is why we need to talk about capacity in kWh AND max charge voltage.

Is there an easy way to establish max charge voltage? I have hunted for TM Spy et al on the UK App Store, with zero success. Do I need a CAN BUS reader of some sort or is there an alternative way. I would love to get the reading as it’s something of a silver bullet.

 

[fbitz777]

So, you say you are 11 miles degraded.
Does that mean your full rated range is 254 (down from 265)?
If so, then you are actually degraded by more because of their change to 276 Wh/mi.
The 265 was based on 295 Wh/mi (78.1 kWh usable)
Your displayed 254 is at 276 Wh/mi (70.1 kWh usable)
So 70.1 kWh / 295 Wh/mi = 237 miles using the same rating method.
265-238 = 27 miles actual degradation (10%)

THIS is why we need to talk about capacity in kWh AND max charge voltage.

You an me seem to be the only two users who have claimed that Tesla changed the Wh/mi rating on our car (mine went from 311 to 285 on X75)...this is BIG as it further reduces the range from the rated range shown on display. The stupid part is they left the rated Wh/mi on the consumption graph (311) while using 285 for the trip planner...

 

[Chaserr]

It's all on Can. With root you can even use Tesla's built in Ludicrous monitor on the IC to see voltage (You don't need to enable Ludicrous to only use the menu option). Tesla has been trying to deny "third party" apps that show their voltage has been capped so they'll probably claim their own data is invalid for some reason, too, but more proof is better and since their legal representation sounds fairly incompetent presenting third party and getting their typical canned response sets them up when they can't say the same about their own official software readings verifying the third party readers.

You an me seem to be the only two users who have claimed that Tesla changed the Wh/mi rating on our car (mine went from 311 to 285 on X75)...this is BIG as it further reduces the range or the car from the indicated rated. The stupid part is they left the rated Wh/mi on the consumption graph (311) while using 285 for the trip planner.

I haven't seen it myself, I'm still 295 - but for users that were manipulated like that it could be dieselgate trouble because since that number comes from the EPA it's not something Tesla can arbitrarily change whenever they want to hide their wrongdoing.

 

[jensk2]

I especially like the guidance of charging to something less than 100% to reduce the impact of the reduction in range..

Long term the guidance could be good. If you look at the '1C Cycle at P1 - P5' graphs in the linked paper, scientist and engineers that measure capacity fade on Carbon Anode based Li-Ion batteries, needs to put in rest durations to perform decent aging measures.

A battery cycled without rest periods can degrade - in their tests - with 15% when cycled around 80% SoC for 300 days. When rested at 80% SoC they only recover a few percent capacity, but if rested at 9% SoC - best case - they recover back to 94% capacity!

The recovery was present when resting at SoC below 45-65% (we do not know the exact upper limit for continued recovery, but the effect is neglible at 65% SoC)

Disclaimer: The paper is for Prismatic Li-Ion with another cathode chemistry than Tesla Model S/X, but both Tesla 18650 NCA Cells and the tested Prismatic cells have Anode Oversizing and Overhang, so mechanism should to some degree be the same)

Ref: 'Investigation of capacity recovery during rest period at different states-of-charge after cycle life test for prismatic Li(Ni1/3Mn1/3Co1/3)O2-graphite cells
Authors:MeinertLewerenzabPhilippDechentabDirk UweSauerabc
https://www.sciencedirect.com/.../pii/S2352152X18306637

(Given the paper is public AND I have given Credit I hope it is okay to show ONE graph from it). The slightly busy graphs shows effect of resting at P1=80%, P2=65%, P3=45%, p4=19% and P5 = 9% SoC after cyling cells around P1, P2, P3, P4 and P5 SoC

Even the Peter Keil thesis listed in other posts in this thread confirm capacity recovery on Panasonic 18650 NCA cells
Peter Keil (All you (n)ever wanted to know about 18650 NCA cells (~ Tesla X+S)
https://mediatum.ub.tum.de/doc/1355829/file.pdf

 

[Ferrycraigs]

It's all on Can. With root you can even use Tesla's built in Ludicrous monitor on the IC to see voltage (You don't need to enable Ludicrous to only use the menu option).

Thank you. Alas my penury S70 can’t even spell Ludicrous far less read it. Indeed it doesn’t even have Insane. Such things are for its bigger brothers. So I shall have to seek another method.

 

[Chaserr]

Thank you. Alas my penury S70 can’t even spell Ludicrous far less read it. Indeed it doesn’t even have Insane. Such things are for its bigger brothers. So that’s not much use for me.

That's why I mention root. You can enable the information display on any car, even if they don't have the ability to use Ludicrous itself.

 

[DJRas]

Is there an easy way to establish max charge voltage? I have hunted for TM Spy et al on the UK App Store, with zero success. Do I need a CAN BUS reader of some sort or is there an alternative way. I would love to get the reading as it’s something of a silver bullet.

I have proposed sending my unit along with a linked cheap smartphone to anyone affected.
It would however be about $40 USD each way to get it to you.

It's all on Can. With root you can even use Tesla's built in Ludicrous monitor on the IC to see voltage (You don't need to enable Ludicrous to only use the menu option). Tesla has been trying to deny "third party" apps that show their voltage has been capped so they'll probably claim their own data is invalid for some reason, too, but more proof is better and since their legal representation sounds fairly incompetent presenting third party and getting their typical canned response sets them up when they can't say the same about their own official software readings verifying the third party readers.



I haven't seen it myself, I'm still 295 - but for users that were manipulated like that it could be dieselgate trouble because since that number comes from the EPA it's not something Tesla can arbitrarily change whenever they want to hide their wrongdoing.

I have data going back to Feb 2019 that shows the 276 multiplier used.