SOC data: 281-based, New-Bars

[rainnw]

If these cells were actually charged all the way to 4.21 volts (aka, true 100%), you would see a bend at the start as well. interesting....

Thats crazy talk, 4.15v is really pushing it.

No, its really not. A lot of lithium ion chargers will actually charge to this voltage per cell. Its a great way to reduce battery life for very little extra capacity.

It looks like the slope at 175 (raw) SOC is only half of the slope below 50 SOC. Therefore TWICE as many miles per bar in the middle than at the low-SOC end.

Lithium ion cells will have an immediate drop, flatten out near their average voltage (3.75, for example) and drop down again rapidly. This makes perfect sense. I am glad to see the "100% SOC" is after this first drop, which indicates that even the 100% SOC is conservative.

 

[garygid]

These are NOT "normal" Li-ion cells. Yes, their voltage drops faster at the beginning and end of the "usable" capacity. The center portion slopes more than many other Li-ion chemistries.

So they are similar, but not the same as other cell constructions.

 

[TonyWilliams]

These are NOT "normal" Li-ion cells.

Here's the data I dug up on the battery from a Nissan press release last year:

Type: Laminated lithium-ion battery
Total capacity (kWh): 24
Power output (kW): Over 90 (the motor is 80kW)
Number of modules: 48 (192 cells)

Battery pack contents:
-Positive electrodes – Lithium manganate
-Negative electrodes – Carbon

Another quip from this forum:
Comparing it to the Chevy Volt battery, " While both packs are basically LiMn2O4, the LG Chem battery does blend in some nickel and cobalt, which presumably we won't see in the Nissan battery until Gen II."

 

[evnow]

Another quip from this forum:
Comparing it to the Chevy Volt battery, " While both packs are basically LiMn2O4, the LG Chem battery does blend in some nickel and cobalt, which presumably we won't see in the Nissan battery until Gen II."

Not correct - Nissan's battery already has some other elements - but first time I've read this about LG.

 

[GregH]

Who out there still gets 281 on a full charge? The last few times I forced a full charge (vs my normal 80%) I ended at 275 or 276.. The most I'd ever seen was 280.

 

[mwalsh]

Who out there still gets 281 on a full charge? The last few times I forced a full charge (vs my normal 80%) I ended at 275 or 276.. The most I'd ever seen was 280.

Not me. Best I've done thus far has been 276 also. Though I'm now beginning to accept the variance as "normal".

Edit: Make that 277.

 

[surfingslovak]

Who out there still gets 281 on a full charge? The last few times I forced a full charge (vs my normal 80%) I ended at 275 or 276.. The most I'd ever seen was 280.

That would correspond to a capacity variation of 1.5-1.8%. I hope you don't mind a few questions. Does this depend on ambient temperature? I've noticed that you took delivery recently, but was this number higher in the past?

 

[garygid]

Does anybody know how to get their LEAF to do more/some/many Equilization cycles?

After the battery is well equalized, does the 280/281 at "full" become easier to achieve.

For now, we have these possible reasons for "full" readings around 275:

1. the battery might be un-ballanced (and charging must stop because the highest cell-pair is "full", but the pack voltage is not at its equalized maximum)

2. the battery might have diminished capacity (but, it is not clear that a lower capacity would show up in this "SOC" value, especially if it is pack voltage-based)

3. some temperature effect (but that is unlikely a factor for mwalsh)

 

[mwalsh]

Does anybody know how to get their LEAF to do more/some/many Equilization cycles?

I'd like to know that too!

 

[garygid]

Hint:
The LEAF can slow-discharge "high" cells for about 50 minutes almost any time, but the following "boost" ("sip" of e-fuel) can ONLY happen if the LEAF can still get AC Power to "charge" for a few minutes more.

My Maple LEAF almost always stays plugged in long after 100% or 80% charging is initially "complete", allowing it to equalize-and-sip if it wants to.

If unplugged (or outside an EVSE's timer window), the LEAF cannot take a "sip" to top-up after the slow-discharging has been done.

 

[Boomer23]

Who out there still gets 281 on a full charge? The last few times I forced a full charge (vs my normal 80%) I ended at 275 or 276.. The most I'd ever seen was 280.

I'm getting 281 on a full charge. I almost always charge to 100% and the car remains plugged in, on timer, for several hours after completing the charge. My TED graph plots don't appear to show any "sipping" after the charge completions.

 

[Herm]

That would correspond to a capacity variation of 1.5-1.8%. I hope you don't mind a few questions. Does this depend on ambient temperature? I've noticed that you took delivery recently, but was this number higher in the past?

It probably does if the 281 number is solely based on pack voltage.. I dont feel confortable using voltage as the sole indicator of SOC.. at least put some temp compensation in.

 

[DaveinOlyWA]

i am guessing ambient temps may play a role. hotter temps might lower the max charge level

in reality, the difference is pretty small and could easily be attributed to tolerance errors on measurement especially if consistent. too bad we dont know what it was fresh from the factory.

 

[garygid]

1. Those reporting the several minutes of "sipping" saw/graphed it in the 45-60 minute window after the charging was "complete".

2. Getting 281, you probably have well-equalized cells, at least at the top end, so no slow-discharge is done, and thus no "top-up" sipping is needed.

2. Some Li-ion battery types (like LiFePO4) actually GAIN a little capacity throughout the first 50 or 100 (or so) use cycles.

 

[surfingslovak]

Reading through this thread, it seems that cell equalization is important and for that you have to charge the battery to full and let it sit for about an hour while plugged in. Is that correct?

I'm not sure if you have seen the Roadster battery care guide by Dan Myggen. I understand that these are two different vehicles, but there are some conceptual similarities, and battery pack balancing is one of them.

http://www.teslamotorsclub.com/showthread.php/3848-Tesla-Roadster-Battery-Care

 

[garygid]

What another car/BMS system does is educational, but only tells us that Battery-Cell Equilization is considered to be important to that car's designer.

There are some who claim (for their purposes/batteries) that cell-ballancing is not necessary.

I think it is necessary, in any real system expected to be in use for years.

It does not tell us what the LEAF does for Battery Management.

From the LEAF Service Manual, we have a reasonable guess about the LEAF's Equalization strategy: Top-Equalize by discharging selected "higher" cells for a while, then (at least in some curcumstances) "sip" a little e-fuel to top-up.

However, we do not (yet) know WHEN (or under what conditions) the LEAF does an Equilization cycle.

If it is ONLY done after a 100% charge, then ALWAYS charging to 80% would NEVER Equalize.

Somebody who can watch/plot their charging or battery current, for at least an hour after the "normal" charging is complete, will begin to tell us WHEN the "sipping" is done.

Using the (slightly-modified) SOC-Meter (and a PC) as a EV-CAN buss logger, should allow us to Log and graph Battery Pack Current.

 

[GregH]

It is certainly much easier to balance at 100% SOC than 80% as cell imbalances stand out that much more near full. Perhaps some of the folks with data logging on the EVSE side or on the vehicle side can illuminate the balance/sip cycle and if it ever happens on an 80% charge? Also for those logging on the vehicle side I'd be curious to know the initial and final RAhC/SOC# during this process.. Has anyone ever seen the pack voltage go beyond 394.5V (4.1V/cell) during charge?

Interesting that the Tesla "storage" mode keeps cells at 20-50%!! That reinforces all I've heard about avoiding time spent at elevated SOC or temperature for optimal cell longevity.. But 20%?!?! sheesh...

 

[DaveinOlyWA]

Another thing to keep in mind that equalization could happen and not change the SOC. we are not changing the pack voltage. Only individual cells which may not be enough to register the very tiny loss associated with the transfer of power from one cell to another

 

[TomT]

I regularly charge to 80 percent and have a TED, and I have never seen any evidence of balance/sip at that charge level. I do occasionally see it on those rare times that I charge to 100 percent.

It is certainly much easier to balance at 100% SOC than 80% as cell imbalances stand out that much more near full. Perhaps some of the folks with data logging on the EVSE side or on the vehicle side can illuminate the balance/sip cycle and if it ever happens on an 80% charge? Also for those logging on the vehicle side I'd be curious to know the initial and final RAhC/SOC# during this process.. Has anyone ever seen the pack voltage go beyond 394.5V (4.1V/cell) during charge?

 

[garygid]

Discharging a few (5) cell pairs by 0.02 volt each will drop the pack voltage by only 0.1 volt.

However, dropping 40 cells by 0.01 volt each is 0.4 volt drop in the pack voltage. Compared to a voltage swing of about 1 volt in each of 96 cell-pairs (96 volts) from usable-full to usable-empty, the 0.4 volt is over one SOC-tick, and it could even show as a loss of two SOC ticks.