Is Charging to 100% Really that Bad?

[stopcrazypp]

Actually, according to the blog post below back in 2006 Tesla was treating a 100% SoC as 4.15v/cell and a 90% SoC as 4.10v/cell. They also treated a battery as having a 0% charge at 3.00v/cell which they say is about a 2% SoC.

A Bit About Batteries

That was for the Roadster. In the Model S (other than the software limited versions like 40kWh and latest 60kWh), 100% is 4.2V.

 

[stopcrazypp]

Your 2015 Leaf doesn't have a 80% charge setting? Weird, my 2013 did. Maybe your 100% isn't a full charge but a 90%ish so you are good to fully charge all the time?

Nissan, in their infinite wisdom, eliminated the ability to charge to less than 100%.

It's not wisdom. The EPA rules averages distinct charging modes for determining range (80% is 66 miles, 100% is 84 miles, average is 75). That's why the 2013 Leaf was rated at 75 miles of EPA range, even though it gets 84 miles fully charged.
EPA Downshifts Its Electric Car Range Formula

In the 2014/2015 model year, Nissan eliminated the 80% charging mode, which changed the EPA range to 84 miles. The EPA rule is one of the stupidest rules ever, but it is what it is. Tesla got around it by allowing a continuous toggle (not distinct charging modes), so the EPA rule didn't apply to Tesla.

 

[stopcrazypp]

It seems fairly accepted on the forums that charging to 100% and letting the car sit for more than an hour should be avoided as it can cause faster degradation over the long term to the battery.

Charging to 100% (other than the impact from the likely deeper cycle) doesn't have a whole lot of impact, but the amount of time and temperature it sits at or near 100% is what matters.

I quantified it in a post somewhere, but couldn't find it, but using the NREL data from an old post, here is the storage loss from sitting at the given SOC for 12 hours a day at 25C for 8 years:
20% SOC: 3.6
40% SOC: 4.2
60% SOC: 5.2
80% SOC: 6.6
90% SOC: 7.9
100% SOC: 9.5
Is a standard charge significantly worse than an 80% charge?

 

[sdorn]

Charging to 100% (other than the impact from the likely deeper cycle) doesn't have a whole lot of impact, but the amount of time and temperature it sits at or near 100% is what matters.

I quantified it in a post somewhere, but couldn't find it, but using the NREL data from an old post, here is the storage loss from sitting at the given SOC for 12 hours a day at 25C for 8 years:
20% SOC: 3.6
40% SOC: 4.2
60% SOC: 5.2
80% SOC: 6.6
90% SOC: 7.9
100% SOC: 9.5
Is a standard charge significantly worse than an 80% charge?

Wow, at 8 years that loss is almost insignificant even at 100%. The car is literally sitting at 100% fifty percent of the time.

 

[stopcrazypp]

Wow, at 8 years that loss is almost insignificant even at 100%. The car is literally sitting at 100% fifty percent of the time.

Keep in mind that this is only the percentage loss attributable to that 12 hours per day of storage. Your cycling loss with also play a part (and that varies based on miles traveled). These numbers are intended to show the relative impact in setting your max charging rate at various SOCs. It's still measurable, but perhaps not as significant as people may expect.

 

[tls]

One data point is the Tesloop limousine service which runs between LA and Las Vegas daily. They have an S 85D and Supercharge it to 100% daily. They've put on 200,000 miles in one year and only noted 6% battery degradation.
Impressive.

Yes, but their usage pattern almost certainly has them reaching 100% charge and then immediately discharging the battery by driving.

This is not like leaving your car in your garage hooked up to a charger that is maintaining it at 100% overnight, every night. Unless I had to, that I definitely would not do.

Here's something I don't understand: charge rate has an impact on battery capacity too -- so, if you're at 50% and you're going to go on a long trip on Monday morning, is it better to trickle up to 100% at low amperage starting Sunday afternoon, or to schedule an 80A charge and blast up there early Sunday morning -- or just get up an hour earlier and hit a Supercharger to put the range on?

Not wanting to spend long periods at high SoC would suggest to put the charge on using the maximum amperage possible, as close to time of use as possible. So, the 80A or Supercharger option. But not wanting to charge at high rate (which also degrades capacity) would suggest to use the lowest feasible charging rate, so, the overnight trickle to get you from 50% to 100%.

With the BMS maintaining battery temperature, and the fast-charge effects largely driven by cell temperature, I bet the answer's somewhere in the middle -- but this is a very tricky optimization problem in at least three independent variables. Someone like wk057 or Ingineer might have the data to head towards a real answer, but generating that data would require sacrificing capacity or even cells on a real battery; it'd be great to know what Tesla thinks.

 

[mspohr]

Yes, but their usage pattern almost certainly has them reaching 100% charge and then immediately discharging the battery by driving.

This is not like leaving your car in your garage hooked up to a charger that is maintaining it at 100% overnight, every night. Unless I had to, that I definitely would not do.

Here's something I don't understand: charge rate has an impact on battery capacity too -- so, if you're at 50% and you're going to go on a long trip on Monday morning, is it better to trickle up to 100% at low amperage starting Sunday afternoon, or to schedule an 80A charge and blast up there early Sunday morning -- or just get up an hour earlier and hit a Supercharger to put the range on?

Not wanting to spend long periods at high SoC would suggest to put the charge on using the maximum amperage possible, as close to time of use as possible. So, the 80A or Supercharger option. But not wanting to charge at high rate (which also degrades capacity) would suggest to use the lowest feasible charging rate, so, the overnight trickle to get you from 50% to 100%.

With the BMS maintaining battery temperature, and the fast-charge effects largely driven by cell temperature, I bet the answer's somewhere in the middle -- but this is a very tricky optimization problem in at least three independent variables. Someone like wk057 or Ingineer might have the data to head towards a real answer, but generating that data would require sacrificing capacity or even cells on a real battery; it'd be great to know what Tesla thinks.

We don't know what their charging pattern is. . They might be charging to 100% every night so they would be ready to go the max distance in the morning

 

[AB4EJ]

Regarding battery charging in general - there seems to be a lot of folk wisdom on this topic - some based on experience, some based on intuition; but I wish I could see some based on a controlled experiment done by a battery manufacturer, where representative samples of batteries are charged & discharged various ways, and the best regimen established by a scientific study (has such a study been done & results published in a peer-reviewed journal?)

One of the things that makes it challenging to do a really good study is that the battery is an electrochemical system that changes over time just due to aging (in addition to whatever charge/discharge pattern is used). But for a test to complete in a reasonable length of time (less than 15 years!) - you have to use accelerated aging to simulate that. (It's tricky to know for sure if your accelerated aging method really represents actual aging). Another complication is that battery chemistry is constantly being improved, so that whatever we figure out today as the ideal charging regimen may be different when new battery chemistry is rolled out.

All that to say, I expect that advice on this topic may be good or not, it's hard to know unless it came from Panasonic or LG, or other battery manufacturer.

 

[tls]

We don't know what their charging pattern is. . They might be charging to 100% every night so they would be ready to go the max distance in the morning

I think we do know at least half of the pattern, since they are hitting a Supercharger in the middle of the trip.

 

[wdolson]

Regarding battery charging in general - there seems to be a lot of folk wisdom on this topic - some based on experience, some based on intuition; but I wish I could see some based on a controlled experiment done by a battery manufacturer, where representative samples of batteries are charged & discharged various ways, and the best regimen established by a scientific study (has such a study been done & results published in a peer-reviewed journal?)

One of the things that makes it challenging to do a really good study is that the battery is an electrochemical system that changes over time just due to aging (in addition to whatever charge/discharge pattern is used). But for a test to complete in a reasonable length of time (less than 15 years!) - you have to use accelerated aging to simulate that. (It's tricky to know for sure if your accelerated aging method really represents actual aging). Another complication is that battery chemistry is constantly being improved, so that whatever we figure out today as the ideal charging regimen may be different when new battery chemistry is rolled out.

All that to say, I expect that advice on this topic may be good or not, it's hard to know unless it came from Panasonic or LG, or other battery manufacturer.

I posted this near the beginning of this thread:
http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

Battery University is an online resource put up there by Cadex who make battery management hardware. They don't make batteries, but they know a fair bit about managing battery systems.

The actual battery manufacturers don't publish all that much on the lifetime and aging of their batteries, they don't want the competition to use it against them. Battery research is one of the hottest things at the university and independent lab level right now. Understanding how they age is a critical part of developing any new battery chemistry. There are some Li-ion chemistries that have great energy density, but they degrade very quickly. These haven't made it into production because not too many people would buy a Model S P150D that required replacing the battery pack every six months.

Accelerated aging testing is an art form all in itself and is never as accurate as real world aging, but in many cases it can tell you some worst case scenarios.

 

[apacheguy]

I normally charge to 75%, but I have no qualms about charging to 100% when I need it. It's not bad at all to let it sit at 100% for a few hours either in my experience. Just don't let it sit there for half a day or longer. Folks who say you must immediately drive it as soon as the charge indicator reaches 100% are going over the top and overthinking this.

 

[FlatSix911]

Li-ion batteries are really a class of batteries that use small particles of lithium trapped in a matrix for one of the terminals and some other material for the other. The matrix is mostly graphite, though silicon can hold more ions and using a mix of graphite and a little silicon is how Tesla/Panasonic boosted the energy capacity of the cells for the 90 KWh/70KWh battery packs.

The chemistry that Tesla uses is rated at about 4.2V at max charge and the nominal voltage (the voltage the cell is for most of the discharge) is 3.6-3.7V. The exact number differs depending on the source. The actual voltage is on a curve that drops from 4-4.2V to around 3.8V quickly as the battery drops from 100% to 90%, then levels out and drops slowly from 90% to about 10% from about 3.8V to around 3.4V, then it drops off a cliff at the end. I wrote more about battery tech here:
Battery Tech Part 2 | Tesla Blitherings

Great blog!

 

[wdolson]

Great blog!

View attachment 200657

Thanks.

It hasn't gotten much attention lately. The last 6 weeks or so have been nuts, no time to write.