Battery Losing Range at Low SOC

[Samo23]

Remaining range at low SOC is always an estimate in any EV.. Some companies hide the imprecision by providing a larger bottom buffer.

My 2017 chevy Bolt does a better job

 

[borski]

My 2017 chevy Bolt does a better job

Cool. Enjoy it in good health.

My 2022 Air DE tracks the battery percentage just fine. At low SOC, any vehicle, including your Bolt, is estimating. Your Bolt just gives you a larger bottom buffer because they don't want you going to some chevy forum and complaining about the fact that an estimate is an estimate.

A battery's open-circuit voltage (OCV), which is often used to estimate SOC, is less reliable at low voltages. Additionally, the battery's discharge rate can change significantly as it approaches a "dead" state, making it difficult to predict remaining capacity. OCV is a measure of the voltage a battery has when no current is flowing. Another method for estimating SOC, coulomb counting (measuring the amount of charge flowing in and out), can also be affected by factors like self-discharge and cell imbalances, making it less reliable at low levels.

Basically, Lucid can either lie to you (like Chevy does), or admit that estimating at low SOCs is very difficult, and impossible to do accurately, due to, well, physics. Lucid chose to give you access to all (or most) of the battery, rather than artificially telling you you have less range than you actually do, but the caveat of this is that the real-time estimation means it may get it wrong.

The solution to all this is don't run your car to zero, because how much charge you have left on the way to zero is a moving target, in every single EV on the planet.

 

[Rogue]

I went from 5% to 1% overnight after a 7.5 hour drive. Was the conclusion from this thread that it should be brought up to customer service?

Considering it myself since this is the first time even at very low states of charge this has occurred.

 

[borski]

I went from 5% to 1% overnight after a 7.5 hour drive. Was the conclusion from this thread that it should be brought up to customer service?

Considering it myself since this is the first time even at very low states of charge this has occurred.

It’s worth having service check battery diagnostics to make sure nothing is wrong. They can check that quickly.

If diagnostics show the battery is fine, this is likely due to the fact that low SOC estimations are immensely difficult to make accurately.

 

[Adnillien]

I went from 5% to 1% overnight after a 7.5 hour drive. Was the conclusion from this thread that it should be brought up to customer service?

Considering it myself since this is the first time even at very low states of charge this has occurred.

SOC estimation error accumulate as you drive and recharge. 7.5 hours is not that long of a drive but it depends on how you were driving and how many times you had to recharge. AS Borski says, ask service to take a look.

The SOC estimate is only accurate when the battery has been resting for several hours with no load and a constant temperature. In this situation, the SOC is determined by voltage and temperature. While very accurate, its still an estimate becasue the voltage and temperature will vary from cell to cell due to differences each cell's capacity and temperature will vary based on where in the module it is located. While driving, the battery is not a rest so the temperature corrected voltage is not very accurate. To overcome this, the BMS system tries to count Coulombs (current) into and out of the battery. This measurement has errors and those errors accumulate over time until the battery rests overnight. Repeated fast acceleration followed by high regeneration along with DC fast charging can increase the SOC error. My guess is that the 1% SOC in the morning was accurate and the 4% error accumulated over the 7.5 hr drive.

 

[L-Dude]

It is mind blowing to me that someone would run the car down so low so often. I'm not criticizing, it is your car to do what you want with it. But I would never run my ICE car down to the last 32oz of gas and expect an accurate reading from my "distance remaining" instrumentation. JMHO YMMV.

 

[Samo23]

yea.... I understand @borski. I appreciate the detailed explanation.

Does losing 20 miles in 30 minutes when car is at idle constitute normal function?

 

[Samo23]

It is mind blowing to me that someone would run the car down so low so often. I'm not criticizing, it is your car to do what you want with it. But I would never run my ICE car down to the last 32oz of gas and expect an accurate reading from my "distance remaining" instrumentation. JMHO YMMV.

No, not often. But when I get on a long drive, there are circumstances when I'd rather go eat than sit at another charger for 30-45 minutes.

This is when this happens. Getting back in my car with 11 miles to go, 35 miles stated range, and getting back in my car with 7 miles left. Unacceptable. Has also happened as picture show from going to get groceries and coming back outside to go charge and wallah 15 miles evaportate in 3.5 hours. That's approx 4kwh that just disappeared

 

[Samo23]

It’s worth having service check battery diagnostics to make sure nothing is wrong. They can check that quickly.

If diagnostics show the battery is fine, this is likely due to the fact that low SOC estimations are immensely difficult to make accurately.

Lucid looked at the cell balancing and determined either BMS/battery is faulty, so. Awaiting what steps come next.



You all can sit down now

 

[DeaneG]

Lucid looked at the cell balancing and determined either BMS/battery is faulty, so. Awaiting what steps come next.

Makes sense. Hope they fix it quickly for you.

 

[borski]

Lucid looked at the cell balancing and determined either BMS/battery is faulty, so. Awaiting what steps come next.



You all can sit down now

Awesome

 

[Steveinarizona]

I think ChatGPT is correct. I recently had a very similar experience with my AGT. Noticed that my range had suddenly diminished from where it normally was, especially when I was at a low SOC. . I explained to my SC that it was as if electrons were "leaking out of my battery". Indeed, when they checked it, cells were out of balance. They replaced my entire battery with a refurbished one. Problem was solved.

Not to steal the thread but if Lucid used a refurbished battery, it might make sense for you to check if it meets this capacity threshold from the Lucid OEM warranty:

"If it is determined that your battery requires warranty repair, Lucid will repair, replace, or adjust the part with a new, reconditioned, or remanufactured part at the discretion of Lucid. However, Lucid will ensure that the capacity of the battery will be equal or greater than the original battery capacity during the warranty claim."

To me, refurbished means you have a reconditioned battery.

 

[borski]

Not to steal the thread but if Lucid used a refurbished battery, it might make sense for you to check if it meets this capacity threshold from the Lucid OEM warranty:

"If it is determined that your battery requires warranty repair, Lucid will repair, replace, or adjust the part with a new, reconditioned, or remanufactured part at the discretion of Lucid. However, Lucid will ensure that the capacity of the battery will be equal or greater than the original battery capacity during the warranty claim."

To me, refurbished means you have a reconditioned battery.

They didn’t. They use a remanufactured battery.

 

[CWedge411]

I just had this happen! Yesterday, I had 85miles of range. Went to grab breakfast 35 miles on the dot round trip.

Get home with 50 miles of range...Check, Spot on!
Didn't go anywhere yesterday because it was raining cats and dogs, but in the 60s temp wise.

This morning i get up and was like "ah crap i forgot to plug it in, but ehh i got 50 miles left to get get breakfast"

Open up my app 13miles left!! WTF! it was in the 50s all night so it wasn't a cold issue. Anyway, Breakfast was 12 miles round trip, made it back with 1 miles remaining and now it's charging, but what the actual F happened to 37 miles of range?

 

[borski]

I just had this happen! Yesterday, I had 85miles of range. Went to grab breakfast 35 miles on the dot round trip.

Get home with 50 miles of range...Check, Spot on!
Didn't go anywhere yesterday because it was raining cats and dogs, but in the 60s temp wise.

This morning i get up and was like "ah crap i forgot to plug it in, but ehh i got 50 miles left to get get breakfast"

Open up my app 13miles left!! WTF! it was in the 50s all night so it wasn't a cold issue. Anyway, Breakfast was 12 miles round trip, made it back with 1 miles remaining and now it's charging, but what the actual F happened to 37 miles of range?

Please read the rest of the thread.

 

[WhistleKing30]

Just stumbled across this thread. Not sure if its still an issue but a bad cell could indeed explain the issue. A battery pack consist of thousands of cells. When a cell is around 4.25V its considered full (100%) while its considered empty somewhere around 2.7V to 3V (0%) depending on the chosen bottom buffer. Anyway, the issue with SoC state estimation is that the decline from 100% to 0% is not linear. Which makes accurate SoC estimations complex - to the point that it cannot be measured accurately but to some degree has to be guessed.

Now the problem: if there's one bad cell amongst the thousands, then the discharge characteristic of this one cell changes. Often this becomes especially apparent at a lower SoC. Bad cells tend increase their internal resistance disproportionally when they're no longer under load, or when the battery pack's temperature change. Thus making the estimation even more difficult. So when you park your car and the pack cools down, this could be a logical consequence in case of a bad cell.

Long story short: my money is on a bad cell.

 

[Samo23]

Just stumbled across this thread. Not sure if its still an issue but a bad cell could indeed explain the issue. A battery pack consist of thousands of cells. When a cell is around 4.25V its considered full (100%) while its considered empty somewhere around 2.7V to 3V (0%) depending on the chosen bottom buffer. Anyway, the issue with SoC state estimation is that the decline from 100% to 0% is not linear. Which makes accurate SoC estimations complex - to the point that it cannot be measured accurately but to some degree has to be guessed.

Now the problem: if there's one bad cell amongst the thousands, then the discharge characteristic of this one cell changes. Often this becomes especially apparent at a lower SoC. Bad cells tend increase their internal resistance disproportionally when they're no longer under load, or when the battery pack's temperature change. Thus making the estimation even more difficult. So when you park your car and the pack cools down, this could be a logical consequence in case of a bad cell.

Long story short: my money is on a bad cell.

Was a bad cell.

But....don't let your car get below 20% ever and don't trust anything below 20% is the advice from this forum. Essentially, 1/5 of your battery is entirely untrustworthy and that's ok

 

[DeaneG]

Was a bad cell.
...Essentially, 1/5 of your battery is entirely untrustworthy and that's ok

If you happen to have a bad cell.

 

[Samo23]

If you happen to have a bad cell.

Nobody knew I had a bad cell when recommending and suggesting haha

 

[segbrk]

But....don't let your car get below 20% ever and don't trust anything below 20% is the advice from this forum. Essentially, 1/5 of your battery is entirely untrustworthy and that's ok

I'd say more like 10%. Which, yeah, the gauge in an ICE vehicle is also a wild guess below that level as fuel sloshes around and expands and contracts with temperature and air pressure. I drove 10+ miles on a fully "empty" tank in my 2015 Mercedes, but it just as easily could have gone the other way. Same with EVs. There's a Lucid range test video on youtube where the gauge said 0% and the driver kept going for miles.
I'd also suggest ignoring the "miles remaining" completely and looking at %. The miles remaining is a guess in general (in every car), and not something you want to be counting down to single digits (or even 10 or 20).