Capacity Loss on 2011-2012 LEAFs

[Herm]

You guys wanted a hard number, now you have it.. 5 hours at 110° F, straight from the horse's mouth.

 

[jspearman]

You guys wanted a hard number, now you have it.. 5 hours at 110° F, straight from the horse's mouth.

I guess all the dealers in the Phoenix metro area will have to pull their cars off the lot during June, July and August, just to be safe. They can build a special air-conditioned Leaf corral.

Seriously, though, I'm 100% sure my personal car sat on the dealer's lot in 110 degree temps, since it was over 110 the day I bought the car, and it was an orphan that had been there for a while. Maybe the dealers here didn't get the Perry memo.

 

[GRA]

You guys wanted a hard number, now you have it.. 5 hours at 110° F, straight from the horse's mouth.

Considering the numerous statements by Mr. Perry which have proved to be inaccurate and/or flat out wrong, are you sure you've got the right end of the horse?

 

[tbleakne]

I submit that we have a problem with the LEAF's pack in very hot climates, and we're seeing predictable range loss in other regions. The degree to which the latter is a problem relates directly to each owner's range needs and whether they are concerned about resale value loss, which may well IMHO be somewhat self-induced due to our own lumping of the predictable range loss in with the heat related problem.

I'm certainly not lumping, as I have plainly said my former LEAF's 15-18% loss in 15 months had nothing to do with temperature, and specified that the car has never been exposed to any "high" temperatures of any kind, ever. I also submit that I don't think that is the norm, or that my car was going to "level off", as is seemingly popular to say. It actually appeared to be in a free fall from even a month ago.
>>>

I also have no confidence that our loss will "level off" as many seem to expect, and I feel we don't have enough data to support this belief. I would not be surprised if there is only a modest slow-down in the loss rate over the next year. The theory says shelf life loss goes as (time)^.5, but it is not clear to me how the accelerated loss due to elevated temperature is related to the shelf life loss process. The temperature may instead be accelerating the loss that is proportional (linear) in battery cycles.

For the past two months I have been trying to mitigate my accelerated temperature loss by aggressively limiting my nightly charge to stop at only 55-65%, and to store the car at 45% SOC while I am away for short trips. I think this measure, only available to me because I don't commute to a job, is helping some, but having no direct readout of battery pack temperature makes it hard to see a correlation.

I have begun to record a daily estimate of what I call my "6B/5B" ratio R. R=0 means the car's battery temperature was at 5 bars for all of the previous 24 hrs, and R=1 means it was at 6 bars for the full 24 hours. I don't yet have enough of this data to display, but in the meantime here is my loss data. "Day 0" is May 30, 2012.

The linear fit yields:
6.0 days per Gid of loss
5.0 Gids lost per month
10.7% SOC loss per 6 months.

The loss of capacity at 100% charge will be greater. I now do them extremely rarely, which is becoming a hardship for some of my outings.

I estimate that the battery temperature will reach 6 bars for some portion of the day for the 6 months of May through October.

 

[tbleakne]

Edit: duplicate post deleted. Sorry.

 

[evnow]

Phx in the summer the lows are 105f and 118 day time temps

I'm guessing you are joking ...

 

[smkettner]

Inside the garage might hold 105 overnight. But I would be parked outside.

 

[DaveEV]

I also have no confidence that our loss will "level off" as many seem to expect

Your theory forgets to take into account ambient temperature. It's already been shown that GIDs will go down as the average temperature goes up and will recover somewhat as the average temperatures goes down. Your plot covers a period of time which warms up significantly in Claremont, CA. The average high has climbed 5F from May through July this year from 82F to 92F (using Ontario Airport as data source).

If the 60 day plot you are showing were a trend, your current GID count would be much lower since ~231 is what it should read when new - yet you only lost ~6 GIDs on an 80% charge in the first year of ownership, but now have apparently lost 10 GIDs in 60 days. Sorry, does not compute.

 

[TonyWilliams]

I honestly don't think he temperature corrections for charging work quite that way. I think it's more like charging amps.... Constant at 120 amps until 50% charge, then taper off on schedule.

I think temp corrections are probably some "one size fits all" until some significant temp, then it begins ramping down the amount of Gids.

 

[Boomer23]

but in the meantime here is my loss data. "Day 0" is May 30, 2012.

The linear fit yields:
6.0 days per Gid of loss
5.0 Gids lost per month
10.7% SOC loss per 6 months.

For what it's worth, tbleakne, my Gids at 80% charge are identical to yours over the same time period, at least the starting and ending numbers for your plot are identical. It's quite striking. As you know, I charged to 100% most of the time for the first 13 months while you have been VERY careful with charging, my car was delivered more than two months before yours, and ambient temperatures where I live are around 12 degrees cooler in the summer than at your house. Yet the Gid counts are behaving the very same way. To me, that smells of shelf life effect dominating, overshadowing temperatures and charging behavior here in So Cal. My theory is that our battery packs were from the same batch and they have been aging in the same way since their "birth".

Here are my Gids at 80% charge beginning June 1, this year:
223, 223, 223, 223, 222, 222, 221, 221, 221, 221, 220, 221, 220, 220, 219, 217, 217 (June 29)
221, 217, 217, 218, 218, 217, 216, 216, 216, 215, 215, 215, 213, 214, 213, 212, 212, 212 (July30)
212, 212 (August 3)

 

[leafkabob]

Anybody get their car back from Casa Grande today? I know they kept a couple of them til Monday, but surely somebody got their car back today. Anything to report?

 

[tbleakne]

I haven't lost a bar yet, but it's right around the corner. I've noticed significant range loss, an 80% charge only gets me 9 bars, and the GOM only displays 75-80 with 100% charge. My driving habits haven't changed, and still average 4.1.

19K miles
S/N: 909
Delivery Date: 4/18/11

I may be running in the same heat (doh! no pun intended) as you.
Received my first 9-bar 80% charge this afternoon.
19.8k mi, days short of 14 mos.

For those that have lost the first bar, what were your Gid readings at 80% at that time?

JP, sorry to hear this. Some additional data about your case would be helpful:

With your OC home location, I would expect that your LEAF shows 5 temperature bars most mornings. Is this true?
After parked at AQMD, is your car at 6 temp bars for ride home in the Summer?
With charge available at both work and home, I would think you rarely need to charge to 100%. Your question above also implies you have more experience at 80%. Is this true ?

If your answers to the above questions are yes, this, together with your relatively high mileage would suggest that your loss is at least partially related to battery cycle count and temperature, and not just temperature and time. If so, I would put your case in the same class as 91040 and possibly TonyWilliams. The last I heard, 91040 was down to about 85% on the Gid meter for 100% charge.

 

[SanDust]

I also have no confidence that our loss will "level off" as many seem to expect, and I feel we don't have enough data to support this belief. I would not be surprised if there is only a modest slow-down in the loss rate over the next year. The theory says shelf life loss goes as (time)^.5, but it is not clear to me how the accelerated loss due to elevated temperature is related to the shelf life loss process. The temperature may instead be accelerating the loss that is proportional (linear) in battery cycles.

For cycle loss lithium cells will lose capacity faster for the first few hundred cycles. After that the curve flattens. It's reasonable to assume the AESC cells will follow the same path.

To me the issue has been made worse by Nissan not being more clear that the early years would see the largest losses. I blame Mark Perry for this. He was too busy hyping to be explaining. When you say 20% capacity loss over 8 years people are going to think the loss is linear, a couple of percent a year, not 10% the first year.

For people focused on temperature and bars, my Leaf has been at six bars for several weeks, during which time I don't think it's gotten to 74F more than one or two days.

 

[tbleakne]

I also have no confidence that our loss will "level off" as many seem to expect

Your theory forgets to take into account ambient temperature. It's already been shown that GIDs will go down as the average temperature goes up and will recover somewhat as the average temperatures goes down. Your plot covers a period of time which warms up significantly in Claremont, CA. The average high has climbed 5F from May through July this year from 82F to 92F (using Ontario Airport as data source).

If the 60 day plot you are showing were a trend, your current GID count would be much lower since ~231 is what it should read when new - yet you only lost ~6 GIDs on an 80% charge in the first year of ownership, but now have apparently lost 10 GIDs in 60 days. Sorry, does not compute.

I am primarily reporting data, not theory. Please don't reject my data. I just displayed a linear fit to limited data. My point is that despite the change in temperatures over these 60 days, the trend looks about linear.

I agree that the lack of detected capacity loss last summer and the relatively sudden onset of capacity loss this Spring does not fit the classic (time)^.5 model. I agree that after I got my Gid meter last Fall I saw some limited improvement in capacity by Winter.

Of course I know the capacity loss is a function of temperature in some non-linear way, but as I stated, I have no way to track the pack temperature accurately. I quoted a 6-month window because I expect the loss to slow or stop this Fall. I also don't know what the thermal time constant of the pack is, but the transitions from 5 to 6 temperature bars and back again show that the time constant is much shorter than 24 hours. This tells me the 24-hr temperature history of the battery is important, not just the average ambient.

 

[RegGuheert]

I also don't know what the thermal time constant of the pack is, but the transitions from 5 to 6 temperature bars and back again show that the time constant is much shorter than 24 hours. This tells me the 24-hr temperature history of the battery is important, not just the average ambient.

I agree that the thermal time constant is less than 24 hours. But I will also state that I think it is a strong function of the speed of the car since the airflow will quickly bring the battery case close to ambient. (The thermal time constants inside the battery case, OTOH, should be much more static. Presumably the thermal resistance from the case to the cells is quite low, which tends to imply a low thermal time constant, also.) The result is that the behavior of the bars is strongly influenced by *when* the car is driven. Specifically, if the car is driven in the hottest part of the day then the temperature of the battery case will quickly go to that level. If the car is simply parked for 24 hours, it just seems to slowly follow the ambient temperature around with about a 4- to 6-hour time constant. It certainly gives me pause to drive the car when it is 80F in the garage and 90F or hotter outside.

 

[mksE55]

Add another TX car to the list. Just the other day lost my 12th bar just at 11,000 miles. I have been following along wondering when I would go since the heat issues looks like an obvious battery problem for Nissan. luckily on a lease but was hoping to purchase the car and cont to save thousands on gas. I am a little shocked I didnt make it to the 12,000mile mark. I purchased the car with 3000miles on it from the dealer, and no they were not using the 80% charge rule when I test drove it or weeks later when I purchased the car. I did a mix of 100% charging in the winter and usually 80% summer. Still have saved over 250.00 monthly in gas. My drive distance is short about 15-25 miles a day so I dont think I have to change my driving habits yet. I will sit back and see what Nissan does about this issue for now.
http://www.flickr.com/photos/84220810@N03/7710447022/

 

[vrwl]

... If your answers to the above questions are yes, this, together with your relatively high mileage would suggest that your loss is at least partially related to battery cycle count and temperature, and not just temperature and time. If so, I would put your case in the same class as 91040 and possibly TonyWilliams. The last I heard, 91040 was down to about 85% on the Gid meter for 100% charge.

The history and limited data from my car might suggest that battery cycle count has nothing (or very little) to do with it. See my signature for additional GID data points.

 

[DaveinOlyWA]

... If your answers to the above questions are yes, this, together with your relatively high mileage would suggest that your loss is at least partially related to battery cycle count and temperature, and not just temperature and time. If so, I would put your case in the same class as 91040 and possibly TonyWilliams. The last I heard, 91040 was down to about 85% on the Gid meter for 100% charge.

The history and limited data from my car might suggest that battery cycle count has nothing (or very little) to do with it. See my signature for additional GID data points.

all of it plays a part. as to the degree i would rate them from worst to best

1) heat
2) time
3) charge cycles

 

[ecoobsessive]

all of it plays a part. as to the degree i would rate them from worst to best

1) heat
2) time
3) charge cycles

Ok, so can someone clarify for me (if we know the answer) which of the above 3 will "level off" and which are linear?


Thanks.

 

[dgpcolorado]

You guys wanted a hard number, now you have it.. 5 hours at 110° F, straight from the horse's mouth.

Considering the numerous statements by Mr. Perry which have proved to be inaccurate and/or flat out wrong, are you sure you've got the right end of the horse?

I'm not sure what Perry's background is, but my impression since last year is that his position [Nissan North America Product Director] with Nissan North America is that of "chief flack". I think his off-the-cuff pronouncements should be taken with a big grain of salt.

If Hidetoshi Kadota were to make a statement about battery life under various conditions, I'd be more inclined to pay attention. Not that it would ever be allowed.