Chevy Bolt EV Forum banner

1301 - 1320 of 1350 Posts

·
Registered
2020 Chevy Bolt and all Tesla models owned by me and my family
Joined
·
461 Posts
That's a huge swing. I wonder if they made it more conservative in the 2020 in cold weather?
I have not done 100-2% battery drain in the winter.
But it definitely acts like Tesla BMC in the winter and reporting much less available Kwh vs Summer.
I can easily manage 3.3-3.7 going 65 MPH in 10°F weather with cabin on manual fan 2 tep. 73°F
Tires if Summer used 40 psi or if winter tires 42 psi
 

·
Registered
Joined
·
8,690 Posts
But that is not very accurate.....any time regen is active it will lower KWH used.
That is just making more complicated to make accurate measure....but i guess it is within 1-5% depending on internal resistance....temperature and etc.
Yes. I have to assume that GM is doing the internal accounting of energy in and out, so should store somewhere, after a complete charge/discharge cycle, how many Ah/kWh were actually available.

It would be pretty trivial to show us that figure. I suppose, from their position, it is much preferable to keep us in the dark. They would get nothing but flack from irate owners every time it changed.
 

·
Registered
2019 LT Slate Grey Metallic
Joined
·
53 Posts
I posted this information in the GM "fix" Announcement Today thread and was asked by @EV Engineering to also post it here, so here goes:

Ok, got some data.
First off, here is a dropbox folder where I will upload my screen snapshots as time goes by:
Bolt2019BatteryRecall
I don't see any PID disappearing after the recall. The 2019 Bat Cap Est is still there!

-1 week before the recall
Battery Capacity Estimated 58.2kWh
Battery Level Displayed 62.7%
State Of Charge HD Raw 63.2%
State Of Charge Raw 63.1%
Battery Cell Voltages Min/Avg/Max: 3.757mv/3.770mv/3.782mv (25mv)

-Evening before the recall
Battery Capacity Estimated 58.5kWh
Battery Level Displayed 74.1%
State Of Charge HD Raw 73.4%
State Of Charge Raw 73.3%
Battery Cell Voltages Min/Avg/Max: 3.871mv/3.879mv/3.887mv (16mv)

-Recall done on 2021-05-21. Charged to 100% first time on 2021-05-24
Battery Capacity Estimated 58.5kWh
Battery Level Displayed 100.0%
State Of Charge HD Raw 95.9%
State Of Charge Raw 96.1%
Battery Cell Voltages Min/Avg/Max: 4.141mv/4.153mv/4.165mv (23mv)

Observations:
After the charge from 75% to 100% was completed (scheduled to end at 1:00pm, ended at 12:59pm). For at least an hour (I unplugged after that), the EVSE didn't shut off, it was still supplying 0.5-0.7A @240V. I didn't hear any battery conditionning going on (garage temps at 22C) and the green light behind the windshield was steady green. I don't recall this behaviour in the past. If I didn't have to leave, I would have let it go on to see for how long. Was it performing some cell leveling or some new diagnostics, I'm not really sure.
 

·
Registered
Joined
·
257 Posts
Something I just noticed about the fix. They reset the number of charges. I was at 2,348 before the update. It went to zero immediately after and now sits at 28. Is this a known thing?
 

·
Registered
2020 Chevy Bolt and all Tesla models owned by me and my family
Joined
·
461 Posts
I posted this information in the GM "fix" Announcement Today thread and was asked by @EV Engineering to also post it here, so here goes:

Ok, got some data.
First off, here is a dropbox folder where I will upload my screen snapshots as time goes by:
Bolt2019BatteryRecall
I don't see any PID disappearing after the recall. The 2019 Bat Cap Est is still there!

-1 week before the recall
Battery Capacity Estimated 58.2kWh
Battery Level Displayed 62.7%
State Of Charge HD Raw 63.2%
State Of Charge Raw 63.1%
Battery Cell Voltages Min/Avg/Max: 3.757mv/3.770mv/3.782mv (25mv)

-Evening before the recall
Battery Capacity Estimated 58.5kWh
Battery Level Displayed 74.1%
State Of Charge HD Raw 73.4%
State Of Charge Raw 73.3%
Battery Cell Voltages Min/Avg/Max: 3.871mv/3.879mv/3.887mv (16mv)

-Recall done on 2021-05-21. Charged to 100% first time on 2021-05-24
Battery Capacity Estimated 58.5kWh
Battery Level Displayed 100.0%
State Of Charge HD Raw 95.9%
State Of Charge Raw 96.1%
Battery Cell Voltages Min/Avg/Max: 4.141mv/4.153mv/4.165mv (23mv)

Observations:
After the charge from 75% to 100% was completed (scheduled to end at 1:00pm, ended at 12:59pm). For at least an hour (I unplugged after that), the EVSE didn't shut off, it was still supplying 0.5-0.7A @240V. I didn't hear any battery conditionning going on (garage temps at 22C) and the green light behind the windshield was steady green. I don't recall this behaviour in the past. If I didn't have to leave, I would have let it go on to see for how long. Was it performing some cell leveling or some new diagnostics, I'm not really sure.
RAW SOC should be 98.xx% at display 100% ... if this is your first time doing 100% after quite some time..

Keep checking this PID on my 2020 it will get up to 98-99% at display 100% .
And definitely if you see after car is showing you charging is done and EVSE is still showing that is pulling power leave it and unplug once it shows it is idling. Sometimes my 2020 could do easily 60 minutes after getting fully charged pulling low amperage from EVSE.
 

·
Registered
2019 LT Slate Grey Metallic
Joined
·
53 Posts
Alright, thanks for that info. It had been 5+ months since charging to 100% (not my choice) 😉
Next time, I will try to set my end charge time earlier and let it go until the EVSE stops.
Could it be part of the "fix" to display 100% but effectively stop at 96%? or am I not understanding the difference between the two values?
 

·
Registered
2020 Chevy Bolt and all Tesla models owned by me and my family
Joined
·
461 Posts
Alright, thanks for that info. It had been 5+ months since charging to 100% (not my choice) 😉
Next time, I will try to set my end charge time earlier and let it go until the EVSE stops.
Could it be part of the "fix" to display 100% but effectively stop at 96%? or am I not understanding the difference between the two values?
On 2020 it will never go over 97.37-98% true SOC.
So i assume that they are masking out even bigger buffer from true 100% but they show you on display 100%.
Without more people posting results it is to early to make any statements.
You also have consider that this max charge is influenced by battery temperature....age of battery number charges....Fast DC charging and etc.
Your numbers look good for the age and mileage.
You could do once run from display 100% to
PID SOC 3-5 % And see what is your avaliable KWH.
You will need to stay around 55 MPH to minimize any loss in internal resistance... if you can't do in one day....no problem you can do 2-3 but stay away from hard acceleration and speed over 55 MPH.
 

·
Registered
2020 Chevy Bolt and all Tesla models owned by me and my family
Joined
·
461 Posts
Something I just noticed about the fix. They reset the number of charges. I was at 2,348 before the update. It went to zero immediately after and now sits at 28. Is this a known thing?
That should be good to watch from other what they see.
 

·
Registered
Joined
·
17 Posts
Tesla uses canisters, which has significantly higher surface area compared to the amount of active material. This leads to slightly less energy density per pack, and slightly higher cost (assuming no economies of scale), but significantly better thermal management.
The rest of your post is spot-on - GM/LG are playing a guessing game with cell temperatures floating where they can't see them - but unfortunately they're not alone. Tesla's packs have the same problem and they're playing the same guessing game: the jelly roll inside a cylindrical cell has very low thermal conductivity in the radial direction. You can cool the external shell of an 18650/21700 as much as you want, but if there are 10-20 watts of heat being generated throughout the whole thing, you're going to have a gradient. If the cell is being charged at 5C, e.g. at a 250 kW supercharger, the gradient can be fairly substantial from the surface to the core.

Also, this shows the thermal gradient with with a cylindrical cell cooled along the entire cylindrical surface. Tesla only has cooling channels on every other cell row - so each cell only gets about 1/3 of its circumference cooled. Nobody has a perfect solution, but anything is better than no cooling at all.

The reality is that the difference in fast-charging speed between a Tesla and a Bolt is a matter of cathode+anode chemistry and corporate risk appetite. Tesla gets away with charging at 5C and GM only does 1C; Panasonic's NCA chemistry has about 1/2 the internal resistance (but 2-3x the volatility) when compared to LG's NMC; Tesla is okay with a few battery fires here and there while driving, charging, or parked, but GM loses their minds over 1 in 10,000 cars catching fire while parked. These facts are related.

P.S. The 4680 "tabless" design will only be superior to ordinary cylindrical cells if it's cooled from the top and bottom. Otherwise, it'll be worse - more diameter (coke can vs. Sharpie) means more spiral layers of jelly roll and higher thermal resistance.
 

·
Registered
2019 LT Slate Grey Metallic
Joined
·
53 Posts
2nd charge to 100% yesterday. This time I programmed it to end 2½h before the time I had to leave. It continued doing that 0.5A current draw from the EVSE until I unplugged it.
Numbers look pretty much the same as the last time.

Battery Capacity Estimated 58.528kWh (0.000)
Battery Level Displayed 100.0%
State Of Charge HD Raw 95.9731% (+0.0289)
State Of Charge Raw 96.0784% (0.0000)
Battery Cell Voltages Min/Avg/Max: 4.140mv/4.152mv/4.164mv (23mv)

I will probably update if/when anything changes...
 

·
Registered
Joined
·
8,690 Posts
Tesla is okay with a few battery fires here and there while driving, charging, or parked, but GM loses their minds over 1 in 10,000 cars catching fire while parked.
The Tesla Model S, and Model X had some thermal runaway events unrelated to a crash.

Have you looked at the NHTSA site for spontaneous battery fire complaints in the Model 3, and Model Y? These cars have sold over an order of magnitude more than the Bolt. I challenge you to find six.
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,312 ·
After the charge from 75% to 100% was completed (scheduled to end at 1:00pm, ended at 12:59pm). For at least an hour (I unplugged after that), the EVSE didn't shut off, it was still supplying 0.5-0.7A @240V. I didn't hear any battery conditionning going on (garage temps at 22C) and the green light behind the windshield was steady green. I don't recall this behaviour in the past. If I didn't have to leave, I would have let it go on to see for how long. Was it performing some cell leveling or some new diagnostics, I'm not really sure.
The final fix keeps the computers on and monitoring the battery for up to 12 hours after the charge has completed. During this time, the 12V system's power is supplied by the APM which is powered by the HV system, which gets its power through the OBCM powered by your plugged in L2.

So this is perfectly expected. 120W or so makes sense. It's likely 50W or so for the computers based on what I've calculated in the past, and float charging the 12V will take some power too. The OBCM and APM are both very inefficient at low power, so I would expect significant losses there.
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,313 ·
RAW SOC should be 98.xx% at display 100% ... if this is your first time doing 100% after quite some time..
In my 2017, never that high. It goes from 96 and climbs to somewhere less than 97.5, after which the charge completes and the value resets to 96.0%. Most often it's less than 97 before charge complete and the reset.
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,314 ·
The rest of your post is spot-on - GM/LG are playing a guessing game with cell temperatures floating where they can't see them - but unfortunately they're not alone. Tesla's packs have the same problem and they're playing the same guessing game: the jelly roll inside a cylindrical cell has very low thermal conductivity in the radial direction. You can cool the external shell of an 18650/21700 as much as you want, but if there are 10-20 watts of heat being generated throughout the whole thing, you're going to have a gradient. If the cell is being charged at 5C, e.g. at a 250 kW supercharger, the gradient can be fairly substantial from the surface to the core.

Also, this shows the thermal gradient with with a cylindrical cell cooled along the entire cylindrical surface. Tesla only has cooling channels on every other cell row - so each cell only gets about 1/3 of its circumference cooled. Nobody has a perfect solution, but anything is better than no cooling at all.

The reality is that the difference in fast-charging speed between a Tesla and a Bolt is a matter of cathode+anode chemistry and corporate risk appetite. Tesla gets away with charging at 5C and GM only does 1C; Panasonic's NCA chemistry has about 1/2 the internal resistance (but 2-3x the volatility) when compared to LG's NMC; Tesla is okay with a few battery fires here and there while driving, charging, or parked, but GM loses their minds over 1 in 10,000 cars catching fire while parked. These facts are related.

P.S. The 4680 "tabless" design will only be superior to ordinary cylindrical cells if it's cooled from the top and bottom. Otherwise, it'll be worse - more diameter (coke can vs. Sharpie) means more spiral layers of jelly roll and higher thermal resistance.
I mean we're both right. You make some great points - but don't forget that the aluminum casing on the cells conducts heat decently well. Pouch cells don't have a casing to conduct the heat. Also 1/3rd of the cylinder surface in contact with the cooling plates is significantly more than the Bolt.

A 2170 cell has ~4600mm^2 of surface area along the sides. ~7.15 square inches.3000 cells - 21,450 sqin of surface area. One third - let's call it 7,000sqin.

The Bolt has 11.5sqin per 2 cells - so 1,656sqin in contact. That gives the Tesla about 4.5x as much contact area. Plus, the Bolt's contact is only through the plate, which increases the distance that the thermal transmission is required, further reducing the efficacy.

I'm sure Tesla's new design will have ways to pull the heat from the ends of the cylinder to however the cooling plates are designed. The cells have thicker walls too, which will increase the thermal transmission.

The chemistry that Tesla chose absolutely plays a part. Having a buffer plays a part. The cooling system plays a huge part. Having cylindrical cells plays a huge part.
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,315 ·
2nd charge to 100% yesterday. This time I programmed it to end 2½h before the time I had to leave. It continued doing that 0.5A current draw from the EVSE until I unplugged it.

Battery Cell Voltages Min/Avg/Max: 4.140mv/4.152mv/4.164mv (23mv)
Hmm, 4.164V maximum is about right. Can you grab the voltages at full charge and after left for a while? I mean best would be to log all night long after the full charge has completed if possible. I want to see if they're balancing the cells more aggressively now.
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,316 ·
Something I just noticed about the fix. They reset the number of charges. I was at 2,348 before the update. It went to zero immediately after and now sits at 28. Is this a known thing?
Yes, the memory inside the HPCM2 is lost with the update, which will erase that value.
 

·
Registered
Joined
·
18 Posts
Noobie question please: is the PID on line #70: "-Battery - Pack - Voltage LD" the one that monitors pack voltage in real time? What does the "LD" mean?

thanks
 

·
Registered
Joined
·
552 Posts
Discussion Starter · #1,318 ·
Noobie question please: is the PID on line #70: "-Battery - Pack - Voltage LD" the one that monitors pack voltage in real time? What does the "LD" mean?

thanks
Low Definition - don't use that one. They all track in real time. Use the HD ones instead.
 
1301 - 1320 of 1350 Posts
Top