Wiki Sudden Loss Of Range With 2019.16.x Software

[VT_EE]

Worst fear confirmed. I posted several weeks back that I was concerned that Tesla might say I only needed a software update to FIX my battery. The fix for my battery prior to v9 being available would have been a replacement of module 58 or a full battery replacement.

Here's the proposed fix now:

"This is <name redacted> from Tesla service. You may remember me from the time you were having the vibration on acceleration issue with your Model S. Hope that is all still good.

Well I’m now doing the diagnosis for the battery alert you have. The repair for this is just a software update and then recheck after latest firmware is installed.

I know how you feel about the software update and I’m requesting your permission to update the firmware to 2019.40 The while idea behind the firmware update is to improve existing system. So there will be no harm done to the car due to a firmware update.


Please let me know how you would like to proceed."

You must feel like your playing Russian Roulette.

 

[sorka]

You must feel like your playing Russian Roulette.

Indeed!

 

[sorka]

The tech is still conferring with the service team on if they can do anything differently than only installing software to "fix" the issue. He said that I can retrieve the car and drive it like it is with the battery alert.

If I can get him to unstage 19.40 then I will choose that path for now. I realized there's additional logging I probably want to get like max current at 15 20 and 25%. Currently(no pun intended) it will pull the full 1516 amps at 30%.

I also want to run it down to empty which I've never done before and measure the module voltages. I want to see when it dies if it's even close to empty or if dies at say 10% because 58 can't keep up. If it dies at 10%, then the only way you can mask this with software is making 10% the displayed 0% i.e. capping on the bottom.

I understand now why they cap the top. If you have a module with dead cells, then the remaining cells will get overcharged as compared to the rest of the modules which will accelerate degradation.

I strongly suspect those that are capped aren't capped at just maximum voltage but also at minimum voltage. It would be really asesome if someone who is not effected and doesn't have a bad module could run their car to empty and take a CANBUS snapshot of the module voltages.

Also, if they're not going to void the warranty, at some point module 58 will be beyond even v10s ability to mask and then I can try again. It will probably be when I break down on the side of the road.

 

[Droschke]

I understand now why they cap the top. If you have a module with dead cells, then the remaining cells will get overcharged as compared to the rest of the modules which will accelerate degradation.

Your understanding is the capped cars must have dead cells?

 

[David99]

. It would be really asesome if someone who is not effected and doesn't have a bad module could run their car to empty and take a CANBUS snapshot of the module voltages.

I have done that before I was affected. Nothing dramatic happens. It's not as simple as a certain voltage of one brick hitting a certain number. I had three shut downs of my car. First one induced with CAN bus logging, second one accidental by pushing too hard, third one just today when voltage dropped dramatically. All three situations had different cell voltages. I also didn't see a particularly bad balancing. In general my battery has excellent balanced cells. At 250k miles it's 4 mV.
Sure if one brick is lower than the others, it can cause a shutdown should it hit the bottom of whatever the BMS decides. There are different situations that can trigger a shutdown.

 

[sorka]

Your understanding is the capped cars must have dead cells?

Speculation. This is why I'd like members who have been capped to post their module voltage cell levels at low SOCs.

 

[qwk]

Your understanding is the capped cars must have dead cells?

They have a bad module, which in turn drags the pack down with it. By capping the voltage, it limps things along for a while, until the bad module drags it down more. Same thing happened with roadster batteries, although Tesla replaced the bad sheet in those.

 

[hpartsch]

Just updated from 2019.32.12.7 to 2019.36.2.3.

My 90% instantly dropped 5 miles from 233 (previously) to 228 (post-update)....

 

[DJRas]

Speculation. This is why I'd like members who have been capped to post their module voltage cell levels at low SOCs.

 

[Dave EV]

They have a bad module, which in turn drags the pack down with it. By capping the voltage, it limps things along for a while, until the bad module drags it down more. Same thing happened with roadster batteries, although Tesla replaced the bad sheet in those.

Tesla top-balances the batteries, so there's never any danger of cells going over 4.2V, even with a module that is weaker than the rest. This is why typically cell-to-cell voltage differences are generally smallest when close to fully charged, Tesla's BMS generally does a good job of keeping the pack balanced.

Now there really isn't any way for Tesla to know directly if a number of cells in a module have popped their fuses. But it could infer that by measuring a modules internal resistance. Tesla's BMS has a pretty good estimate of each module's capacity and internal resistance.

 

[Ferrycraigs]

The tech is still conferring with the service team on if they can do anything differently than only installing software to "fix" the issue. He said that I can retrieve the car and drive it like it is with the battery alert.

If I can get him to unstage 19.40 then I will choose that path for now. I realized there's additional logging I probably want to get like max current at 15 20 and 25%. Currently(no pun intended) it will pull the full 1516 amps at 30%.

I also want to run it down to empty which I've never done before and measure the module voltages. I want to see when it dies if it's even close to empty or if dies at say 10% because 58 can't keep up. If it dies at 10%, then the only way you can mask this with software is making 10% the displayed 0% i.e. capping on the bottom.

I understand now why they cap the top. If you have a module with dead cells, then the remaining cells will get overcharged as compared to the rest of the modules which will accelerate degradation.

I strongly suspect those that are capped aren't capped at just maximum voltage but also at minimum voltage. It would be really asesome if someone who is not effected and doesn't have a bad module could run their car to empty and take a CANBUS snapshot of the module voltages.

Also, if they're not going to void the warranty, at some point module 58 will be beyond even v10s ability to mask and then I can try again. It will probably be when I break down on the side of the road.

Not sure I like the sound of any of that. Good luck!

 

[Ferrycraigs]

Just updated from 2019.32.12.7 to 2019.36.2.3.

My 90% instantly dropped 5 miles from 233 (previously) to 228 (post-update)....

I would say that’s unlikely to be capping. The drop is, in comparison to capping drops (around 30 miles normally) very small. I haven’t heard any reports about capping coming from any download other than 2019.16.1.1. If you are already at 32, my money would be the drop is down to another cause. No, I’m sorry, I don’t know which one. It could just be down to winter.

 

[hpartsch]

I would say that’s unlikely to be capping. The drop is, in comparison to capping drops (around 30 miles normally) very small. I haven’t heard any reports about capping coming from any download other than 2019.16.1.1. If you are already at 32, my money would be the drop is down to another cause. No, I’m sorry, I don’t know which one. It could just be down to winter.

My 90% has been very consistent since I purchased the vehicle used 3 years ago and my degregation has been under 1%. I live in WA state -- and our winters here are overall mild (and lack snow for the most part). I've never had a 90% of 228 until now. A very odd set of circumstances -- especially since I just updated yesterday and all of a sudden 5 miles vanish.

 

[DJRas]

My 90% has been very consistent since I purchased the vehicle used 3 years ago and my degregation has been under 1%. I live in WA state -- and our winters here are overall mild (and lack snow for the most part). I've never had a 90% of 228 until now. A very odd set of circumstances -- especially since I just updated yesterday and all of a sudden 5 miles vanish.

It is quite possible that you will continue to lose a few miles every day or two over the next few weeks if you have been capped.

 

[N..8]

The tech is still conferring with the service team on if they can do anything differently than only installing software to "fix" the issue. He said that I can retrieve the car and drive it like it is with the battery alert.

If I can get him to unstage 19.40 then I will choose that path for now. I realized there's additional logging I probably want to get like max current at 15 20 and 25%. Currently(no pun intended) it will pull the full 1516 amps at 30%.

I also want to run it down to empty which I've never done before and measure the module voltages. I want to see when it dies if it's even close to empty or if dies at say 10% because 58 can't keep up. If it dies at 10%, then the only way you can mask this with software is making 10% the displayed 0% i.e. capping on the bottom.

I understand now why they cap the top. If you have a module with dead cells, then the remaining cells will get overcharged as compared to the rest of the modules which will accelerate degradation.

I strongly suspect those that are capped aren't capped at just maximum voltage but also at minimum voltage. It would be really asesome if someone who is not effected and doesn't have a bad module could run their car to empty and take a CANBUS snapshot of the module voltages.

Also, if they're not going to void the warranty, at some point module 58 will be beyond even v10s ability to mask and then I can try again. It will probably be when I break down on the side of the road.

Sorka this was from last year but you can see I was at about 1 mile of charge left. I was using TM-Spy and as you know it adds the 4kWh buffer into its numbers. The voltages were at about 3.1 for the most part, I need to do this again and see what they look like now.

 

[Droschke]

I would say that’s unlikely to be capping. The drop is, in comparison to capping drops (around 30 miles normally) very small. I haven’t heard any reports about capping coming from any download other than 2019.16.1.1. If you are already at 32, my money would be the drop is down to another cause. No, I’m sorry, I don’t know which one. It could just be down to winter.

I thought we saw few owners reporting (even with graphs) unusual gradual drop recently. In other words, instead of a sudden 20-30 miles overnight drop, some cars are going through smaller chunks of range loss which might not be that noticeable to some as Batterygate. If more cars are still being capped, by witnessing the public outrage due to major losses, why not to make it incremental?

 

[Ferrycraigs]

My 90% has been very consistent since I purchased the vehicle used 3 years ago and my degregation has been under 1%. I live in WA state -- and our winters here are overall mild (and lack snow for the most part). I've never had a 90% of 228 until now. A very odd set of circumstances -- especially since I just updated yesterday and all of a sudden 5 miles vanish.

I agree it sounds odd. I’m just saying there are many reasons for a (small) drop in range. Capping is certainly one. But I think, in this case, it’s unlikely to be the culprit. That doesn’t mean it’s not, just unlikely to be (IMO).

I hear you when you say you are not a new owner, that your 90% range has been consistent and that your winters are mild. I live at 56° North, a little further North than the 49th parallel, and also in a climate that is mild, by which I mean seldom sub zero, and normally snow free. But here winter (temperature) has a noticeable effect on range, in both my, and my wife's EV. In her car it results in a lower range figure on the dash (100 miles down to about 80 miles, so about a 20% drop). In my MS I use % so I have no idea what the reported winter range figure is. (I don’t even know if it reports a smaller figure in winter - others may know). I accept that doesn’t prove much other than it’s just another factor that MAY be responsible. Of course it could be that Tesla have changed the consumption algorithm figure and so produce a lower figure, (but I would have expected them to change it the other way). The only way to confirm if you have been capped is to get the kit to interrogate your BMS, pull some bits of the car apart, connect the special cable to the TDC, connect a Bluetooth OBD II transmitter to an App that reads the data. But from what you say, I wouldn’t think it was worth the expense or hassle. As many of us have found, even proving we have been capped holds no ice with Tesla.

 

[Ferrycraigs]

I thought we saw few owners reporting (even with graphs) unusual gradual drop recently. In other words, instead of a sudden 20-30 miles overnight drop, some cars are going through smaller chunks of range loss which might not be that noticeable to some as Batterygate. If more cars are still being capped, by witnessing the public outrage due to major losses, why not to make it incremental?

Sadly I now believe anything is possible from Tesla. I’m absolutely not discounting this as a possibility.

 

[Droschke]

But here winter (temperature) has a noticeable effect on range, in both my, and my wife's EV. In her car it results in a lower range figure on the dash. In my MS I use % so I have no idea what the reported range figure is. (I don’t even know if it reports a smaller figure in winter - others may know).

The car displayed range (when charged to 90%, for example) is not affected by temperature changes. The actual consumption (wh/mi) is.

 

[jkennebeck]

Bloomberg - Are you a robot?

Not sure if this bodes good or ill for us....


edit: also not sure why the link posted are you a robot...
headline is TEsla loses its 3rd Gen Counsel in the course of a year