what hurts more: charging 100% at all, or time spent there?

[ljwobker]

This is probably a more generic question to battery technology than the LEAF, but here goes. If this is answered somewhere else a pointer would be great - I couldn't figure out what to search on.

I'm trying to figure out if either of these cases gives meaningfully different battery life... (somewhat simplified math, I know the charging times aren't exactly right but that's not the point so ignore it)

case 1: you come home at 8pm with the battery at 20%. You plug it in, and by midnight it's charged to 100%. It spends 8 hours charged at 100%, before you drive it off at 8am. You've had one "80% cycle".

case 2: you come home at 8pm with the battery at 20%. You plug it in but with a timer, so by 11pm it's charged to 80%. The car then spends 8 hours charged at 80%, and you re-start the charge at 7am and by 8am it's charged to 100%. In this case, you have still gone through the same cycle - one trip from 20%->100% charge. But the difference is that you've only spent 1 hour above 80%, instead of 8 hours in case 1.

So the real question is whether case 2 provides meaningfully longer battery life than case 1. Does anyone have real data on this behavior?

 

[TomT]

Scenario 2 is the best.

 

[ljwobker]

data? references? evidence?

 

[ebill3]

Try this: http://goo.gl/ysNE4m

A couple of snips from the article:

Always maintaining a Li-ion battery in a fully charged condition will shorten its lifetime. The chemical changes that shorten the battery lifetime begin when it is manufactured, and these changes are accelerated by high float voltage and high temperature.

Avoid charging to 100% capacity. Selecting a lower float voltage can do this. Reducing the float voltage will increase cycle life and service life at the expense of reduced battery capacity. A 100-mV to 300-mV drop in float voltage can increase cycle life from two to five times or more.

And, this from Tesla: http://www.teslamotorsclub.com/showthread.php/3848-Tesla-Roadster-Battery-Care

Yes, I realize the first reference is about small consumer cells and not large automotive cells, but the chemistry is similar. A little googling will reveal the same facts - storage at high levels of charge is not good.

Note that while 4.2 volts per cell is considered to be the maximum charge for Li cells, neither Nissan nor Tesla ever allow their cells to get to 4.2 volts.

 

[johnrhansen]

Now if you get home at 9pm and you are sure that you won't need the car for the rest of the night, yeah charging it as late as possible is better, but personally, I never allow my car to sit at home at 20% charge. I immediately get the charge up to at least 50 and let it sit there... What if something happens and you need your car? If you have a old truck laying around just in case, OK... but if not, plug your car in for an hour or so to get the charge up a bit, and then let the charge finish in the middle of the night. If you forget to unplug it now and then, too bad so sad.. It won't kill the car to sit at full charge once in awhile... Especially at these winter temperatures!

 

[Nubo]

Now if you get home at 9pm and you are sure that you won't need the car for the rest of the night, yeah charging it as late as possible is better, but personally, I never allow my car to sit at home at 20% charge. I immediately get the charge up to at least 50 and let it sit there... What if something happens and you need your car? If you have a old truck laying around just in case, OK... but if not, plug your car in for an hour or so to get the charge up a bit, and then let the charge finish in the middle of the night. If you forget to unplug it now and then, too bad so sad.. It won't kill the car to sit at full charge once in awhile... Especially at these winter temperatures!

It would be nice if the timers were robust enough to handle this automatically. Or at least a way to choose target SOC when you press timer override.

 

[johnrhansen]

yeah, I just do it manually. You could control the pwer to the evse with a commercial motor contactor that is activated by a programmable household light timer, but it's easier for me to just go out to the garage every time. I have a short commute, and it's not very often I get to 20 percent anyway.

 

[ljwobker]

Someone here showed how to build a webpage where you could one-click do things like turn on/off climate control and start charging. So I'm going to just use a scheduled task (windows) or a cron job (unix) to start charging each morning at a specified time.

 

[apvbguy]

why I lease, I charge it, I drive it or let it sit and guess what, it's a none issue because in 15 months they get the car back

 

[apvbguy]

And, this from Tesla: http://www.teslamotorsclub.com/showthread.php/3848-Tesla-Roadster-Battery-Care

Yes, I realize the first reference is about small consumer cells and not large automotive cells, but the chemistry is similar. A little googling will reveal the same facts - storage at high levels of charge is not good.

Note that while 4.2 volts per cell is considered to be the maximum charge for Li cells, neither Nissan nor Tesla ever allow their cells to get to 4.2 volts.

are your cites really relevant to the battery in the LEAF?

 

[ebill3]

And, this from Tesla: http://www.teslamotorsclub.com/showthread.php/3848-Tesla-Roadster-Battery-Care

Yes, I realize the first reference is about small consumer cells and not large automotive cells, but the chemistry is similar. A little googling will reveal the same facts - storage at high levels of charge is not good.

Note that while 4.2 volts per cell is considered to be the maximum charge for Li cells, neither Nissan nor Tesla ever allow their cells to get to 4.2 volts.

are your cites really relevant to the battery in the LEAF?

Yes.

 

[johnrhansen]

I say it is an issue because these batteries have an environmental cost in their manufacture. Just like everything else, the less that is wasted, the better off everyone is.

why I lease, I charge it, I drive it or let it sit and guess what, it's a none issue because in 15 months they get the car back

 

[GregH]

Leave it at 20% as long as possible and set the timer to charge to 100% in time for your morning departure so that it spends as little time as possible at 100%. Hours spent at 110% (4.2V, impossible on a Leaf) is bad. Hours spent at 100% (4.1V) is kinda bad. Hours spent at 80% (4.05V) is relatively bad. Hours spent at 50% (3.9V) is not bad but not great either. Hours spent at 20% (3.7V) would be optimal (for the life of the battery). Hours spent under VLBW (<3.6V) probably not great...

 

[pkulak]

why I lease, I charge it, I drive it or let it sit and guess what, it's a none issue because in 15 months they get the car back

I'm 90% sure I'm giving this car back in 2 years, but maybe not. I might as well keep the battery in good shape. What if gas is $8/gallon by then and electric cars are worth a fortune/impossible to find? There is nowhere near enough production capacity to keep up if something like that happens. Even used Prius prices spike when gas goes up, and Toyota is churning those out like crazy.

 

[RegGuheert]

So the real question is whether case 2 provides meaningfully longer battery life than case 1. Does anyone have real data on this behavior?

Case 2 should provide a longer battery life. Whether or not that difference is meaningful depends on many other factors which will shorten the battery's life. In other words, if your battery only lasts one year due to hot climate and high mileage, then it likely makes very little difference. But if you live in a cold climate and drive very little, then this might change the battery's life by a year or more. But proving that this this is so is virtually impossible simply from field data, even once we have years more of it.

data? references? evidence?

Please read through the Wiki linked at the top of this page for more details and lots of references. Stoaty has done an outstanding job of compiling a huge amount of information on this topic that he has collected over the years from discussions on this site.

My understanding is that at high charge levels, the high electric field strength causes the electrolyte to break down more quickly. Assumedly the high field strength moves some ions in closer proximity, thus accelerating some undesireable non-reversible reactions which take place in this type of cell. Data that I have seen, including some on the Wiki, indicates that charging to 80% really gives no longer life versus 100%. But I believe storage at about 30% should have some benefits.

 

[xado1]

i bought my leaf for 45.000.-$ cash here in austria,80% QC to 96%,i do not care that much for the battery,i am expecting a better and bigger battery in 2-3 years,then i will buy the bigger one,and sell the original battery.

 

[Stoaty]

Leave it at 20% as long as possible and set the timer to charge to 100% in time for your morning departure so that it spends as little time as possible at 100%. Hours spent at 110% (4.2V, impossible on a Leaf) is bad. Hours spent at 100% (4.1V) is kinda bad. Hours spent at 80% (4.05V) is relatively bad. Hours spent at 50% (3.9V) is not bad but not great either. Hours spent at 20% (3.7V) would be optimal (for the life of the battery). Hours spent under VLBW (<3.6V) probably not great...

I haven't seen any evidence that 20% SOC is better than 50% SOC in any meaningful way. In spite of keeping my Leaf at around 30% SOC for the vast majority of its life, my Leaf is almost exactly matching the Battery Aging Model:

Predicted Loss: 16.46%

Actual Loss: 16.91%

Do you have any references or links backing up that assertion?

 

[ljwobker]

Stoaty, RegGuheert : the wiki is gold. I'm off to read through it. Thanks!

 

[GregH]

Leave it at 20% as long as possible and set the timer to charge to 100% in time for your morning departure so that it spends as little time as possible at 100%. Hours spent at 110% (4.2V, impossible on a Leaf) is bad. Hours spent at 100% (4.1V) is kinda bad. Hours spent at 80% (4.05V) is relatively bad. Hours spent at 50% (3.9V) is not bad but not great either. Hours spent at 20% (3.7V) would be optimal (for the life of the battery). Hours spent under VLBW (<3.6V) probably not great...

I haven't seen any evidence that 20% SOC is better than 50% SOC in any meaningful way. In spite of keeping my Leaf at around 30% SOC for the vast majority of its life, my Leaf is almost exactly matching the Battery Aging Model:

Predicted Loss: 16.46%

Actual Loss: 16.91%

Do you have any references or links backing up that assertion?

Just talking to the battery chemistry guys I've known over the years.. It's well established that higher voltages and especially higher temperatures are worse. The question then becomes how cold is too cold and how low a voltage is too low a voltage. While batteries might preserve well at 0C they're not all that practical or efficient due to the high impedance. As for ideal storage voltage I've seen plenty saying that 3.6V is good but nothing saying 3.9V is preferable. Certainly 3.9V is better than 4.1V, but is it worse than 3.6V? My take had been that the internal stress on the battery was directly related to the higher voltages. From that you might gather than 3.2V was better than 3.6V (for storage) but I don't know that and I've heard others talk about other degradations that happen at low voltages (although usually under 3.0V, hard to reach in a Leaf). Due to the weakness of cells at low voltages I certainly am as gentle as possible with my pack when under VLBW, but I don't have any evidence to suggest there is any problem storing a Leaf AT VLBW (ie, around 3.6V/cell), and it may actually be better than storing at 3.9V (50%)

I'm not suggesting driving at low voltages, merely if it's a bad thing to let the car rest at low voltages for a time before charging. Even I instinctively put a little back in if I arrive home well under VLBW (close to turtle), but I don't know that's necessary and I certainly have no reason to believe (other than vehicle readiness) that there is any advantage to bringing a 3.6V cell up to 3.9V ASAP. It may even be detrimental. Although probably several orders of magnitude less so than leaving it sitting at 4.1V or at an elevated temperature for extended times.

I don't lose any sleep over leaving my Leaf at VLBW for extended periods before charging. It may actually be beneficial. Perhaps some battery chemistry guys on MNL can clarify this.

 

[ljwobker]

Is there a way to add data points into the spreadsheet or help the community improve tracking info? I have a 2012 LEAF that's got ~13k miles and ~15 months on it, and my degradation (Assuming the android app correctly reports Amp-hours/capacity correctly) also is extremely close to the spreadsheet model. 57.75Ahr/87.17% capacity.

Do we have enough data to strongly suggest that leaving the car charged closer to 100% has little impact? For the first year of ownership I left it on 100% charge and plugged in virtually whenever it was not driven, suggesting that it spent more time at a high SoC than would be "ideal" ... but if the incremental degradation isn't even modeled in the spreadsheet AND there are lots of vehicles that don't show up way outside the model, doesn't that imply that the time spent at high SoC isn't really meaningful (and we should all just leave our cars at 100% charge all the time?)