Deducted 5% of SOC almost immediately after parking

[ResHacker]

Hi guys, another thing which I have experienced twice so far and would like to discuss is above described phenomenon. When it occurred for the first time, my brother was driving long trip to Amsterdam and arrived with low SOC 5%. He was quite late there, so he planned to solve the necessary stuff and then charge on nearby SC. But what little bit shocked them was the 0% SOC he saw when entering the car back again. Nevertheless, car behaves like normal, it didn't show any yellow intermittent line on the power meter which I instructed him to watch after.. And without any problems he drove to the SC and charged. Charging speed was also normal except it behaves exactly like the SOC is 5% shifted (eg. on 50% SOC PWR only approx 95kW while normaly is 105kW and so.. ).
Few days back I have experienced the same behavior with the only difference the car used to have 12% when I parked it for a while and later then it turned to 7%.
The question is, why BMS does that math as it happened only in quite short period after the arrival, so the battery was not much colder or so..
Still quite weird behavior in my eyes..

 

[mswlogo]

Doesn’t surprise me at all. When you are that low, you are playing with fire.

Don’t trust any reading below 10%.

If you goose the throttle below 5% it could disappear in a moment.

It’s an estimate based on a voltage.

 

[ucmndd]

This happens to me regularly if I park with a low SoC (say 15% or less) within an hour or so.

It’s normal. The car even tells you as much when you park - “there will be less energy available as the battery cools, we recommend you charge now” or similar message on the MCU.

 

[ResHacker]

Doesn’t surprise me at all. When you are that low, you are playing with fire.

Don’t trust any reading below 10%.

If you goose the throttle below 5% it could disappear in a moment.

It’s an estimate based on a voltage.

Well, sure it's not healthy for the pack and also might be risky, but when I sometimes manage to do so, I monitor the battery via SMT tool + also watch the eventual occurring yellow intermittent power limitation sign which clearly tells you, there's some limit or no. Bjorn also used to check it while doing his range tests.
And also another thing is, if you don't sometime discharge the pack down to let's say a single digit of SOC, the BMS doesn't have enough data to provide you regarding SOC, so then you definitely shouldn't trust it. But I believe that if you do so once upon a time, the BMS will safely know, how much is still left.
And last thing, I wouldn't say it has something to do only with voltage. Sure, voltage is one of the factors, which has influence to the math behind SOC, but definitely not the only one.

This happens to me regularly if I park with a low SoC (say 15% or less) within an hour or so.

It’s normal. The car even tells you as much when you park - “there will be less energy available as the battery cools, we recommend you charge now” or similar message on the MCU.

Yes, so it seems to be a common behavior. It is also clear that when this "SOC shift" happens, I can then charge to like above 100%, so it really doesn't make a sense to me.
Sure, if I arrive in winter with hot battery and it then gets cold within few hours, some % will be deducted, that totally makes a sense to me as cold battery couldn't provide that amount of energy as warm. But I have also experienced, if battery is cold and during the ride it gets warm, the SOC was not changing till 20km driven. That's also clear and makes a sense. But deducting 5% SOC directly after arrival when SOC is low is still weird to me.
I believe it has something to do with scarying the driver and to push him to charge ASAP, it has nothing to do with reality of the current SOC.

 

[RedY23]

And also another thing is, if you don't sometime discharge the pack down to let's say a single digit of SOC, the BMS doesn't have enough data to provide you regarding SOC, so then you definitely shouldn't trust it. But I believe that if you do so once upon a time, the BMS will safely know, how much is still left.

Do you have solid knowledge of this or just an opinion? Respectfully asking because I want to know if your comment comes from fact or forum talk/deduction? Its important to know before guys start to drain the battery. Makes sense what you are saying, but somehow I doubt its true. I think the system looks at battery voltage and at a certain number shows you ZERO. There is no memory or learning.

My comments are based on comments from AAKEE on another forum chat.

 

[ResHacker]

Well, it's not only my opinion. You can also find that info on various web-sites informing how to maintain a Tesla battery like this one. And it makes a big sense to me as otherwise the BMS would have the data only from the SOC spectrum you're using - meaning specific cell voltages <=> SOC. Therefore if I let it from time to time go down to single digit %, I'm not that much worried as it's just good not to push it hard when doing so and being gentle

 

[KArnold]

My '17 MS100D has 80k miles with the expected degradation. Yesterday I drove about 200 miles stopping at a SC on the way for the last 100 miles. Charged enough to have an estimated 12% upon arrival. Driving speeds were 70 to 75 MPH. Sure enough, I had the indicated 12% on arrival.

After an hour or so parked the battery said 5%. By morning it was 1%. Obviously no sentry was on. Ambient temperature was about 50f on arrival dropping to 30f overnight. I've done the cell rebalance recently and on several occasions in the past.

My question is, when driving do I really have 12% or closer to something much lower? Is that simply a matter of the battery cooling and would be relatively accurate while driving? Is it dangerous to plan on arrival with only a 12% buffer?

TY in advance.

 

[ucmndd]

My question is, when driving do I really have 12% or closer to something much lower?

The car is "really" at 12% SoC but because of the cold temperatures it won't let you access all of that energy and will shut down to protect the battery. So in effect, you shouldn't count on that energy being available.

Is it dangerous to plan on arrival with only a 12% buffer?

If you can't charge right away and it's going to be relatively cold, then yes.

 

[KArnold]

The car is "really" at 12% SoC but because of the cold temperatures it won't let you access all of that energy and will shut down to protect the battery.

Only after parking/cooling I assume? But I could have potentially driven further as the car warms the HV battery as needed?

TY again.

 

[ucmndd]

Only after parking/cooling I assume?

Right.

But I could have potentially driven further as the car warms the HV battery as needed?

Maybe? But again, the car isn't going to let the battery provide that energy when it's cold - for heating, propulsion, or otherwise. So if you don't have an external source to provide energy to heat the pack, you can't really count on the idea that the pack will heat itself to give you back some of that range.

As a rule I'd not park the car overnight in cold temps with less than 20% SoC.

 

[Zuikkis]

There was already discussion about this in an earlier thread:

BMS weirdness

So yesterday I had a lot of driving to do, so started out the morning with a 100% charge. I finished the day back home at about 8% charge. I needed to do some work in the garage, so had to park the car on the street for about an hour, maybe 2, before plugging it in inside the garage. When I...

teslamotorsclub.com

There I posted an image which kind of explains this..

This is temperature vs capacity graph of Panasonic NRC18650PD cell, which is not exactly the cell Tesla uses, but probably is close.

You can see that to get the maximum "advertised" capacity the cells need to be at 40C. Any cooler and the capacity starts to drop considerably.

If you could heat the cells you could get the range back, but obviously there's no easy way to do that without external power sources (charging)..