BEWARE! Model S died at 20% charge

[whitex]

Couple of things that might explains this:

1) Cold weather
2) really miscalculated SOC gauge
3) Bad battery

Or all 3.

This car is a 60 it seems from the picture so that is a bad sign to begin with. 60s don't do well in cold due to low capacity. Moreover, this is an old 60s so it only gets worse. I suspect this car is fine in high temperature, but once you get cold weather, it might be too old to keep a high enough voltage to prevent the BMS from tripping. Hopefully Tesla takes this car out of service or at least upgrade the battery when loaning out in cold weather.

It could also be a v9 bug. If it was never fully tested with S60's and they forgot 60 battery specific battery safety shutdown information (which includes environmentals like temperature) so it's shutting down based on some default voltage levels probably from 85, or something like that.

I am not saying that is it, but a possible #4 for your list.

All that said:

Tried everything we could to get the car to accept the charge while it was still on the flatbed. It would not accept the charge. Tow guy called Tesla again, they were no help.

This to me says the car was just broken - something died in the battery or associated systems.

 

[EVSteve]

Guys, this is definitely not a miscalibration issue. SOC is determined by voltage and that is an absolute measurement.

Rated range and battery capacity are calibration variables but not SOC.

SOC is determined by voltage however unbalanced cell groups can cause sudden unexpected drops in output. This risk is increased in cold weather and on older higher mileage cars with smaller batteries where the buffer to shutdown voltage is less. Hence why all cars should periodically charge to 100% so the BMS can peak the individual cells in each group and account for anomalies when predicting remaining SOC.

 

[whitex]

SOC is determined by voltage however unbalanced cell groups can cause sudden unexpected drops in output. This risk is increased in cold weather and on older higher mileage cars with smaller batteries where the buffer to shutdown voltage is less. Hence why all cars should periodically charge to 100% so the BMS can peak the individual cells in each group and account for anomalies when predicting remaining SOC.

Yea, but that shouldn't stop the car from charging (at least BMS should not allow the cells to be so unbalanced that you can't charge the battery anymore).

 

[scole04]

How many miles were on the s60 loaner?

 

[GlmnAlyAirCar]

Guys, this is definitely not a miscalibration issue. SOC is determined by voltage and that is an absolute measurement.

Rated range and battery capacity are calibration variables but not SOC.

Not at all true. Range is directly proportional to SOC by a constant factor.

 

[2012MS85]

@ivan801 As I recall, you received one of the very first MR 3s off the line and delivered to you. For what type of service issue(s) is your 3 in the service center? Sorry you went from a brand new Tesla to having to deal with this broken S60 I pick up my MR 3 tomorrow in Chicago and am hopeful they've worked out the bugs

 

[Akikiki]

I don't get it. I read the OP's post that started this thread. But I don't understand what he's trying to make us aware of with his Beware?

 

[BerTX]

12V battery dead.

All this speculation is fun, but useless. We'll never know what it was -- it wasn't the OP's car.

 

[Tam]

I don't get it. I read the OP's post that started this thread. But I don't understand what he's trying to make us aware of with his Beware?

Lots of people think that they can drive beyond 0% to negative territory but he cautions us that our luck might run out even at 20%!

 

[apacheguy]

Not at all true. Range is directly proportional to SOC by a constant factor.

Not at all true. SOC is fixed to the voltage. Meaning at 3.95 V (cell voltage) my car always displays 75% SOC. Similarly, 4.09 is 90% and 4.20 V is 100%. This is based on years of empirical measurements on my own car using Tesla’s data. I know what I’m talking about.

However, rated range fluctuates and is not tied to SOC by a constant factor as you suggest. Instead it is based on battery capacity as measured by the BMS using output and input energy.

Further reading here:
Calculate usable battery capacity based on rated miles values

 

[apacheguy]

SOC is determined by voltage however unbalanced cell groups can cause sudden unexpected drops in output. This risk is increased in cold weather and on older higher mileage cars with smaller batteries where the buffer to shutdown voltage is less. Hence why all cars should periodically charge to 100% so the BMS can peak the individual cells in each group and account for anomalies when predicting remaining SOC.

The OPs car had 19% SOC. there is no way that one cell group fell below the threshold voltage unless there is something seriously wrong (e.g. a dead module).

Charging to 100% is absolutely not necessary. I haven’t charged my S to 100 in over 6 months. Guess what? The cells are all within 4 mV of each other. This indicates a perfectly balanced pack.

 

[whitex]

Not at all true. SOC is fixed to the voltage. Meaning at 3.95 V (cell voltage) my car always displays 75% SOC. Similarly, 4.09 is 90% and 4.20 V is 100%. This is based on years of empirical measurements on my own car using Tesla’s data. I know what I’m talking about.

However, rated range fluctuates and is not tied to SOC by a constant factor as you suggest. Instead it is based on battery capacity as measured by the BMS using output and input energy.

So what is the SoC if half the cells are at 5% and half at 35% (extreme unbalance)? Is the SoC just a sum of all cell SoCs? If so, you can have 20% overall SoC and not be able to reach sufficient pack voltage because the discharged cells cannot deliver the voltage (would drop too low under load). The extreme unbalance should not happen, but obviously there was something wrong with the battery if it wouldn't accept charge either.

 

[apacheguy]

So what is the SoC if half the cells are at 5% and half at 35% (extreme unbalance)? Is the SoC just a sum of all cell SoCs? If so, you can have 20% overall SoC and not be able to reach sufficient pack voltage because the discharged cells cannot deliver the voltage (would drop too low under load). The extreme unbalance should not happen, but obviously there was something wrong with the battery if it wouldn't accept charge either.

Exactly, there would have to be something seriously wrong with the pack (e.g. dead module) as indicated in my previous post. There is no way that the OPs experience is normal.

Edit: Actually, the proper way to determine SOC is to use the *lowest* cell group voltage since this is ultimately what determines whether the battery can continue to deliver power. Thus, in your example the SOC would be 5%.

 

[EVSteve]

The OPs car had 19% SOC. there is no way that one cell group fell below the threshold voltage unless there is something seriously wrong (e.g. a dead module).

Charging to 100% is absolutely not necessary. I haven’t charged my S to 100 in over 6 months. Guess what? The cells are all within 4 mV of each other. This indicates a perfectly balanced pack.

Absolutely essential to accurate bms calibration especially on older cars. One of many threads demonstrating this to be fact.

2013 P85, 60K miles, 238 max range?

 

[PhilDavid]

Yesterday I dropped off my Model 3 to have some testing done and received a Model S as a loaner. It was a 60 RWD model.

I needed to drive up the canyon for business so after supercharging to 85% I headed out (about 50 miles away). Outside temps were around 10F, so charge was dropping quickly and it did not look like I would make it home on the remaining charge (65 miles away). My backup plan was to top off at a supercharger and then drive home from there (supercharger was in the opposite direction from home, but what are you going to do?). Eight miles from the supercharger I still had 20% left and it was all downhill from there so I was not worried.

All of the sudden I get a warning: "Battery power too low" followed by something like "Pull over now". So I do, and the car stays on for a few minutes, but it does not respond to any commands. Then it just turns off and everything goes black.

After waiting on hold with Tesla for 30 min, everyone I talk to seems to have no idea what they are doing. Instead of calling a tow truck from Evanston, 8 miles away, they call a tow from Park City, 50+ miles away. IT TOOK 2 HOURS for a tow truck to get there. IT WAS 10F (-10C) outside. I repeatedly told them that Evanston was much closer, and that it was very cold.

The tow truck driver could not find where to connect a tow hook, so he called Tesla and the guy on the phone told him to remove a small plastic cap. This was 2014 model. There is no plastic cap, you have to remove the nose cone. After looking for a while, tow guy was about to pop off one of the parking censors when I suggested removing the nose cone.

Got back to Salt Lake, superchargers were down, but we found a regular charger at the service center a mile down the road from supercharger. Tried everything we could to get the car to accept the charge while it was still on the flatbed. It would not accept the charge. Tow guy called Tesla again, they were no help. I just called an Uber and left for home. Got home around 12:30 am.

Would you mind posting the VIN for the car so that if it ends up in the CPO inventory, someone considering buying the car would know about the battery situation?

 

[ivan801]

Hi guys, I am back with some answers to your questions:

- Yes, power consumption was high, it was super cold and I was doing 75 mph.
- There was no way to put the car into tow mode as all the screens were dead and not re-booting.
- I had heated seats but not the heated air for about first 20 minutes. Then everything went dead.
- Not sure if 12V was at fault as the hazard lights worked for 2+ hours and door locks worked as well.
- Yes, I am one of the first to receive 3 MR. Minor repair, they replaced bluetooth module or something like that.
- SC promised to let me know what the issue was, so I will update when they contact me.
- Loaner car had 59k miles on it.

Now for the BEWARE part, as one of you asked why the title:

- Clearly, at least in this case, you can't count that your battery will last down to 0%. There appears to be an issue with some of the Teslas, although I have taken my Model 3 down to 3% with no problem and I am sure that many S60 owners have taken their cars down to 0%.

- Perhaps this can be an additional reminder to all of us to keep warm clothes in the car in the winter. I was extremely cold, shivering uncontrollably towards the end of the two hour wait, even though I had a coat and scarf on. At one point I called the sheriff's office, but they told me it will be 45+ minutes for them to get to me. That was about 1.5 hours into my wait, so tow ETA was faster.

- Depending on what the cause is, perhaps those with the same configuration should have SC check their cars.

- You should not necessarily trust that Tesla will know what they are doing. In my case, instead of waiting on hold for ever, and then waiting for a tow truck from Park City, I should have just called one of the several tow companies from Evanston, only 8 miles away. Even if I had to cover the cost myself, I would have been warm. Towards the end of my wait I would have gladly paid $1,000 for a warm place to wait things out.

 

[Akikiki]

I'm sorry, I stayed out of this as long as my OCB would permit. (Yes, I know I should bite my tongue and keep my mouth shut, but, it didn't happen.) I know better than to never say never. I disagree with the use of absolutes unless a person has all of the facts. I don't have all these facts, but from what I can gleam from what's been written, I think: With such a flag waving title, this thread is likely to get a lot of attention - especially from new owners or prospective owners. And as such its going to be misleading.

In Tesla's defense. It appears that the OP is a recent new owner. I don't see any evidence that would suggest the OP is armed with enough first hand experience or knowledge to make statements like "There appears to be an issue with some of the Teslas, although I have taken my Model 3 down to 3% with no problem and I am sure that many S60 owners have taken their cars down to 0%." Driving one or two or even a half dozen loaners does not quality a person to state "There appears to be an issue with some Teslas" - unless some is actually one.

Running the battery down to 0 is known to risk bricking the battery. There's no recharging from a bricked battery by normal means. OP don't run your battery down to 0 or 3% unless you are willing to risk damaging your car. Continuing to do so, you are going to get some more experience with loaners. Also no evidence or facts to support a statement like "I am sure that many S60 owners have taken their cars down to 0%".

Sorry you got stranded. But you invited fast battery depletion with the - doing 75 mph when it was Super cold, because that hints that you had not given the battery time to have warmed up from being cold soaked. Yes, power consumption WOULD high - that was normal the way you describe what you did. I'm not trying to anger you, but it sounds like you are a victim of inexperience with a Tesla or a Tesla in winter.

Maybe there was something wrong with the battery - yep it sounds like unbalanced (been there). Sounds like it may have been headed towards displaying a problem regardless, but also sounds like it would have gotten you home had you been more familiar with driving a Tesla in cold weather and being easier on the loaner.

Best of luck to you.

 

[whitex]

Exactly, there would have to be something seriously wrong with the pack (e.g. dead module) as indicated in my previous post. There is no way that the OPs experience is normal.

Edit: Actually, the proper way to determine SOC is to use the *lowest* cell group voltage since this is ultimately what determines whether the battery can continue to deliver power. Thus, in your example the SOC would be 5%.

So a single dead cell with very low SoC would make the entire pack show this low SoC. A single dead cell with 0% would make the entire pack show 0%, which I don't think happens, so maybe they don't determine SoC as the SoC of the lowest charged single cell.

 

[diaretical]

Exact same thing happened to me twice. 2013 Tesla P85. Tesla replaced the HV battery under warranty and gave me a loaner battery (which I still have after a year) and didn't charge me a dime. Currently have 151,000 miles and still love the car after 5 years.

 

[Akikiki]

Exact same thing happened to me twice. 2013 Tesla P85. Tesla replaced the HV battery under warranty and gave me a loaner battery (which I still have after a year) and didn't charge me a dime. Currently have 151,000 miles and still love the car after 5 years.

Me too. Same thing. Same year, P85. SC told me after they installed the loaner battery, that mine would be shipped to Fremont, worked on, refurbished and returned to my car. They didn't know when. Six months later, I sold the car and ordered a new one. I checked with SC, they said the loaner battery would become permanent, that there was a policy change and the original would no longer be returned to the car's They had learned what they needed to know. I was good with that. The new battery was charging at 90% two miles higher than my original one was when it was brand new at 90%.