Capacity Loss on 2011-2012 LEAFs

[Roadburner440]

So I have a 2012 Leaf at 32k miles. My AHr is 46.46, SOH 70%, Hx 50.10%, but I still have the 10th bar. What's the deal with this? Shouldn't I have lost the 10th bar a while ago?

I think as well it takes a number of start cycles to drop a bar similar to resetting a check engine light in a gas vehicle. Especially since the ahr can fluctuate up and down depending on temperature, state of charge, etc. I myself am about to drop below 50ahr here in the next week or so I imagine as I continue to cycle my battery deeper as time drags on. Winter will be interesting. I will have enough charge to do my driving and run my heater, but I think it will take me down to 2-4 charge bars where it used to only take me 2 total bars to do my drive. How things have changed in almost 4 years, lol.

 

[edatoakrun]

...I think it will take me down to 2-4 charge bars where it used to only take me 2 total bars to do my drive. How things have changed in almost 4 years, lol.

The actual available kWh remaining in your pack is not discernible from the LBC's gid report, the low capacity warnings, or the lower charge bar displays.

My LEAF's actual kWh capacity available from all these displays (AKA the LEAF's "gauge problem") has increased considerably since delivery.

My pack now has well over 50% of the total (~18.1 KWh warm pack average) capacity available from "100%" to turtle, in only the last four charge bars.

For example, last Thursday afternoon I drove ~59.3 (odometer) miles on the first eight charge bars, and drove another~54.9 miles (~1.6 miles past the VLBW) and still had one charge bar and (I estimate) 14 to16 miles of range left before the turtle.

There was a small average efficiency increase over the course of the trip, due to the ambient temperature increase and slightly lower average speed.

The 112.6 odometer miles to the VLBW was my all-time personal record, made possible by the warm temperatures and the fairly level route from Vacaville to Corning CA.

Not too shabby (If I do say so myself) for A 2011 warm-climate LEAF, ~4.5 years old, with ~40,000 miles, with ten capacity bars, and a ~25% LBC report of ( and I estimate between13% and 14% of actual) capacity loss.

 

[Stanton]

The 112.6 odometer miles to the VLBW was my all-time personal record, made possible by the warm temperatures and the fairly level route from Vacaville to Corning CA.

I might believe that if you were driving downhill the whole way :shock:
What was your miles/kW? I would have been "dead" after your first 60 mile drive!

 

[cwerdna]

What was your miles/kWh? I would have been "dead" after your first 60 mile drive!

FYP. I've noticed you've mixed up the units a bunch of other posts.

kW and kWh are very different metrics. It's the same as confusing gallons with horsepower. Think of kW = horsepower, kWh = gallons.

(BTW, 1 hp = ~0.746 kW and 1 gallon of gasoline=33.7 kWh.)

 

[TomT]

I wonder if that is straight gasoline or Ethanol blend gasoline... Ethanol has less energy so it does make a difference...

(BTW, 1 hp = ~0.746 kW and 1 gallon of gasoline=33.7 kWh.)

 

[DaveEV]

The 112.6 odometer miles to the VLBW was my all-time personal record, made possible by the warm temperatures and the fairly level route from Vacaville to Corning CA.

I might believe that if you were driving downhill the whole way :shock:
What was your miles/kW? I would have been "dead" after your first 60 mile drive!

Yeah, my last drive past VLBW a couple weeks ago was 58 miles at 4.1 mi/kWh. Ended at 20 gids, 9.4% SOC. Ed must be approaching 8 mi/kWh to get 112 mi to VLBW.

 

[edatoakrun]

="Stanton"

The 112.6 odometer miles to the VLBW was my all-time personal record, made possible by the warm temperatures and the fairly level route from Vacaville to Corning CA.

I might believe that if you were driving downhill the whole way :shock:

Actually, there are about 220 ft. of net elevation gain on the entire ~114.4 mile route between the two charge points.

="Stanton" What was your miles/kW?

There are many incorrect ways to answer that question (three of which are below) and one fairly accurate correct answer.

Below are the results for the entire Vacaville (Nut Tree Plaza L2) to Corning (Heritage RV park) segment of my trip:

Actual miles driven/LBC (gid) kWh use report:

~114.4/14.3 = 8.0 m/kWh

Odometer miles driven/CarWings kWh use report (the Nav screen m/kWh display):

114.2/14.8 = 7.7 m/kWh

CarWings miles driven/CarWings kWh use report (the dash m/kWh display):

111.4/14.8 =7.5 m/kWh

Actual miles driven/estimated actual kWh use:

~114.4/~16.5 = ~6.9 m/kWh

="drees"Yeah, my last drive past VLBW a couple weeks ago was 58 miles at 4.1 mi/kWh. Ended at 20 gids, 9.4% SOC. Ed must be approaching 8 mi/kWh to get 112 mi to VLBW.

Looks like you got the answer furthest from the correct one, drees.

 

[DaveEV]

Looks like you got the answer furthest from the correct one, drees.

??? I guessed a number approaching 8 mi/kWh (in my head I was thinking 7+ mi/kWh) and the lowest you gave was 6.9 mi/kWh with the car reporting 7.5-7.7 mi/kWh it would appear that I nailed it. The difference between 6.9 mi/kWh and 8 mi/kWh is only 14% in terms of efficiency.

Actual miles driven/estimated actual kWh use:

~114.4/~16.5 = ~6.9 m/kWh

How are you coming up with this number?

 

[edatoakrun]

="drees"

Looks like you got the answer furthest from the correct one, drees.

??? I guessed a number approaching 8 mi/kWh (in my head I was thinking 7+ mi/kWh) and the lowest you gave was 6.9 mi/kWh with the car reporting 7.5-7.7 mi/kWh it would appear that I nailed it. The difference between 6.9 mi/kWh and 8 mi/kWh is only 14% in terms of efficiency.

If you are content with "only 14%" under-report error in actual kWh use, a gid meter may give you the level of accuracy you desire, but probably only over a full discharge cycle.

I suggest that you stop thinking of the LBC reports and the common kWh use reports from the dash/nav screens and CarWings as interchangeable.

These two data sources report quite different amounts of kWh use as I drive, and the same might be obvious to you, if you do not hold the preconception that they are ~identical.

Where the LBC data really gets ridiculous for my LEAF, is nearer to the bottom of the pack.

I think TickTock was the first forum member to notice wh/gid variability over the entire charge/discharge cycle, over three years ago:

1 gid *mostly* equals 80Wh

http://www.mynissanleaf.com/viewtopic.php?f=44&t=9689

But unless I had conducted a series of accurate range tests over partial charge ranges, I wouldn't have seen how extreme the LBC error-increasing over time- is in under-reporting Wh/gid at the bottom of my LEAF's pack.

As my pack has aged, the actual available kWh at the low end of the pack (as reported by the LBC's charge bars and low battery warnings) has increased, evidently by ~the same error rate as for kWh over the entire charge (~11.5%) as reported by the dash/Nav screen m/kWr and CarWings energy use.

So It seems certain the gids at the bottom of my pack contained, on average, much more than 80 Wh when new, and have increased in average Wh content (evidently by ~11.5%) over time.

For example, my LEAF travelled 25.4 (odometer) miles from the LBW to VLBW on the trip referenced, which the LBC reported as two kWh consumed (from 3.9 to 1.9 kWh remaining, displayed by the LEAF app) or ~12.7 m/kwh, which is probably impossible to get in a LEAF over a ~level route, at any speed.

In fact, due to the factors I already mentioned, I probably was getting slightly better m/kWh in those miles, than over the entire ~114.2 (odometer) mile trip, I'd estimate ~7.1 m/kWh, meaning my LEAF probably actually used ~3.6 kWh to drive those 25.4 (odometer) miles

So, had I reset my nav screen (I did not, as finding that value was not my primary goal) I would have seen ~7.9 m/kWh nominal (7.1x 1.115) on that screen over that LBW-VLB trip segment, as opposed to the 7.7 m/kWh over the trip as a whole.

On this segment of the range/capacity test the 25 gids (49 to 24) from LBW to VLB contained an average of ~144 Wh (3.6 kWh/25 gids) each.

="drees"

Actual miles driven/estimated actual kWh use:

~114.4/~16.5 = ~6.9 m/kWh

How are you coming up with this number?

That's the actual miles driven (114.4) divided by the nav screen/dash/Carwings reported kWh use (14.8) multiplied by 1.115, which = ~16.5 kWh.

I've been estimating my LEAF's available kWh capacity and the under-report factor with range/capacity and recharge capacity tests since ~May of 2014, when I believe my actual capacity loss was probably ~10.5%, while LBC indicated loss was then ~19.5%, with a ~9% error. The rates of both LBC-reported and actual capacity loss seem to have slowed down significantly since then.

I wrote this a few months ago:

...There is considerable variation in the percentage of capacity the LBC allows you to access between "100%" and the lower SOC's, LBW, VLBW, turtle and shutdown, in repeated cycles, so you would want to try to use the largest sample of charge cycles as possible.

The most accurate method, IMO, would be to discharge to the lowest SOC allowed, and meter the recharge to "100%" multiple times over a short interval.

Since it isn't practical for me to fully discharge and monitor the recharge multiple times each month, about a year ago I started to monitor the nominal efficiency of that charge cycle I use the most, from "80%" to ~LBW (and sometimes almost to VLBW in the Winter) which is what I use to make the 50-55 mile trip to the Sacramento Valley floor and back, usually ~ ten times per month.

I use the dash, nav screen and CW energy reports to estimate the nominal kWh error, by finding the nominal discharge/charge cycle efficiency error, which seem to be confirmed by individual range tests...

The nominal kwh use (dash and Nav screen m/kWh) for my six trips "80%" to ~ LBW in June '15 totaled 51.6 nominal kWh (from CarWings daily use reports) while the six recharges back up to "80%" required ~65.2 kWh from the meter, a nominal cycling efficiency of ~79.0%.

Nominal cycling efficiency of only~79.0% (L2, 16 amp, ~240 V) is improbable, IMO,.

As was my LEAF's nominal 6.4 m/kWh overall last month, reflecting the same ~11% under-report of kWh use, meaning my actual average driving efficiency for the month was likely ~ 5.9 m/kWh, dash m/kWh and CarWings both having the same ~2.5% under-report of miles driven, partially offsetting the common ~11% under-report of kWh use.

So I believe my LEAF instead probably used ~57.3 actual kWh (51.6 X 1.11 nominal kWh error) last June which would imply actual cycling efficiency of ~87.7%...

http://www.mynissanleaf.com/viewtopic.php?f=30&t=18269&start=90

 

[agnl]

So what's the average decrease rate in AHr? I posted five days ago that my AHr was 46.46, now it's 46.18. It seems like it's dropping quick and don't get me wrong, I'm all for it, It just seems a bit odd.

 

[DaveEV]

If you are content with "only 14%" under-report error in actual kWh use, a gid meter may give you the level of accuracy you desire, but probably only over a full discharge cycle.

Dude, I took a wild guess at what your dash-reported efficiency must have been to get that kind of range on a 4+ year old California LEAF. 15% is great. I'm not sure what you think I was implying. I can only think that you completely misunderstood the context of my statement.

I suggest that you stop thinking of the LBC reports and the common kWh use reports from the dash/nav screens and CarWings as interchangeable.

Where did I suggest such a thing? In fact, I have recommended against it because of the inaccuracy of both.

These two data sources report quite different amounts of kWh use as I drive, and the same might be obvious to you, if you do not hold the preconception that they are ~identical.

Again, you appear to be inserting words into my mouth.

Where the LBC data really gets ridiculous for my LEAF, is nearer to the bottom of the pack.

I agree that GIDs at least can be significantly non-linear. This has been well documented as you state.

Since it isn't practical for me to fully discharge and monitor the recharge multiple times each month

That is completely unnecessary. All you have to do is periodically charge from turtle to 100% and record the energy from your calibrated meter and compare over time. Your crazy desire to mix multiple partial charges from somewhere around LBW to 80% and mix and match CARWINGS and meter data only introduces more room for error if all you want to do is compare how much energy in your pack you have available from 100% to turtle.

That said, I think that while measuring energy to charge from turtle to 100% probably gives you the best indication of how much capacity you've lost, most people are simply not comfortable driving significantly below LBW, so comparing energy from LBW to 100% and VLBW to 100% is probably a more accurate representation of how much capacity you've lost in actual usable capacity of the vehicle.

So let's look at my most recent LBW to 100% charge data and the first recorded LBW to 100% charge data I have:

2015-08-18: 14.26 kWh
2012-09-30: 16.92 kWh

That's a loss of 16% in 3 years - so I probably lost at least 7% in the previous 6 months for a total of 23% loss of capacity. This matches up well with my seat-of-the-pants estimate of 23% given that I only get about 50 miles from 100% to LBW todayand used to get about 65 miles when the car was new.

How does that match up with the other metrics provided by the LBC?

Pretty close - on Aug 18th LBC reported 49 Ah and 208 GIDs when fully charged. Using 66 Ah and 281 GIDs as new, those numbers suggest about 74% capacity remaining.

Now I know what you're going to say - "but since your example doesn't go below LBW where some extra amount of capacity is stored, the LBC is actually under-reporting capacity by even more!"

This is very likely to be true - but it doesn't matter. As an end user all we have to go off are what the car tells us. And if the car tells us we have lost 23% between 100% and LBW - why should we have to assume otherwise? Nissan had multiple chances to fix the car to do the right thing. And while they may have improved it with one of the firmware updates to the car, it's still wrong.

 

[cwerdna]

So what's the average decrease rate in AHr? I posted five days ago that my AHr was 46.46, now it's 46.18. It seems like it's dropping quick and don't get me wrong, I'm all for it, It just seems a bit odd.

I'm sure that varies a lot depending on temperature. As we've seen, batteries in hot climates degrade very rapidly (e.g. Phoenix and much (?) of Texas) while those in the Pacific Northwest (e.g. Seattle area) degrade a LOT slower.

 

[RegGuheert]

Now I know what you're going to say - "but since your example doesn't go below LBW where some extra amount of capacity is stored, the LBC is actually under-reporting capacity by even more!"

This is very likely to be true - but it doesn't matter. As an end user all we have to go off are what the car tells us. And if the car tells us we have lost 23% between 100% and LBW - why should we have to assume otherwise? Nissan had multiple chances to fix the car to do the right thing. And while they may have improved it with one of the firmware updates to the car, it's still wrong.

This is the real crux of the issue (and many of the disagreements) with the LEAF. The LBC squirrels away more-and-more capacity below LBW while the capacity of the battery gradually drops. It's an unfortunate situation, since it means the car is quite a bit more usable than it appears to be.

Here are some of my thoughts (all SOCs are as reported by LeafSpy):

- For the record, our MY2011 LEAF (built June 2011) has had the P3227 update and after nearly 28,000 miles there is one capacity bar missing and LeafSpy reports about 20% capacity loss.
- LBW has increased from 17% SOC when new to 26% SOC today. That's a 50% increase on a car indicating 20% capacity loss.
- What I see is that the first few charge bars now drop away much more quickly following a full charge than they did when the car was new.
- I currently lose over seven charge bars before the SOC drops below 50%.
- Yesterday I lost eleven charge bars after a 64-mile trip. I returned home with 22.4% SOC remaining. When the car was new, I would have gotten home around LBW. Yesterday, I was five miles past LBW, but probably should have been right at it given the SOC.
- The last two winters have been VERY cold here, and range really does suffer when it is 10F outside. I didn't experience this the first winter of 2012/2013 since it was not as cold.

The bottom line is that the I find I can do much more with the car today than the instrumentation lets on. I'm not likely to benefit from the capacity warranty that Nissan has provided and I want to get as many years out of this battery as I can. As such, I'm becoming more-and-more comfortable with driving below LBW given the information provided by LeafSpy. It's not ideal, but it works. My range is not nearly as limited as it appears on face value!

But I have almost zero experience driving below VLBW since I have only been there a few times. But if the car keeps squirreling away capacity and there is some to be had down there, I may eventually start using that. I appreciate the reports of capacity that Ed makes for driving below VLBW because it gives a glimpse of what may be going on with the LBC.

 

[DaveEV]

- For the record, our MY2011 LEAF (built June 2011) has had the P3227 update and after nearly 28,000 miles there is one capacity bar missing and LeafSpy reports about 20% capacity loss.
- LBW has increased from 17% SOC when new to 26% SOC today. That's a 50% increase on a car indicating 20% capacity loss.

Hmm, that's not much different than my car which is at 48.1 Ah - last LBW I hit occurred at 27% after 53 miles @4.5 mi / kWh. A couple days later I went down to VLBW (14%SOC) after 63 miles @4.3 mi / kWh. So no real capacity hidden between LBW and VLBW for me. But I did get to 67 miles @ 4.3mi/kWh and 8.0% SOC 14 GID. I don't recall how far after VLBW I drove, but with perhaps 8 GIDs left to turtle, I might have had another 3-4 miles range left, so I might have 3-4 miles more miles between VLBW and turtle than expected. Not exactly a huge discrepancy.

- The last two winters have been VERY cold here, and range really does suffer when it is 10F outside.

I'm finding that regen just continues to get worse and worse. With the battery in the upper 70F range I only get 15 kW of regen at 70-80% SOC. Luckily, it doesn't get all that cold here.

 

[RegGuheert]

Hmm, that's not much different than my car which is at 48.1 Ah - last LBW I hit occurred at 27% after 53 miles @4.5 mi / kWh. A couple days later I went down to VLBW (14%SOC) after 63 miles @4.3 mi / kWh. So no real capacity hidden between LBW and VLBW for me. But I did get to 67 miles @ 4.3mi/kWh and 8.0% SOC 14 GID. I don't recall how far after VLBW I drove, but with perhaps 8 GIDs left to turtle, I might have had another 3-4 miles range left, so I might have 3-4 miles more miles between VLBW and turtle than expected. Not exactly a huge discrepancy.

I'm finding that regen just continues to get worse and worse. With the battery in the upper 70F range I only get 15 kW of regen at 70-80% SOC. Luckily, it doesn't get all that cold here.

Which all tends to indicate that SOME of the issues with range are related to battery resistance and how it interplays with the cell balancing circuitry. At high power levels (both charging and discharging), the shunts cannot keep the battery balanced due to the discrepancy in resistance of different cells. This effect shows up clearly when driving on the highway in extremely cold weather: You can see which cells are coldest by looking at which ones have the lowest voltage. Ed's long-distance drives at very slow speeds tell us that much of the capacity is still there, but is is not available to those driving on the highway. In other words, if you are driving at 60 MPH, then Tony's tests are the most important for you and the range will not be significantly different than what the instruments tell you, but if you need to go farther you can use Ed's approach of slowing way down. My driving tends to be somewhere between Tony's and Ed's, so my results also fall somewhere in between.

At the end of the day, I find that the following are necessary to make longer range trips with a degraded MY2011/2012 LEAF:

- The battery must be fully charged and balanced. This gets harder and harder to achieve as the battery ages. Charging at L1 or turning on climate control at the end of the charge can help.
- Drivng slowly can make a significant difference in range, moreso than when the car is new.
- Using LeafSpy to become aware that bars disappear sooner and LBW has crept up (in real terms) and realizing that the same trip that used to not trigger alarms now triggers them much earlier, but you may actually arrive at close to the same SOC as when the car was new.
- In extremely cold and windy whether, we just take another car. The range simply is not there anymore. Two trips last winter where I had to oppotunity charge in very brutal conditions taught me that lesson.

Bottom line: As the battery degrades, the LEAF gets even more fiddly and limited than when it was new.

 

[TimLee]

...
Bottom line: As the battery degrades, the LEAF gets even more fiddly and limited than when it was new.

+1000 :roll:
Maybe Nissan marketing should use that phrase in their next advertisement :lol:

 

[Stanton]

But I have almost zero experience driving below VLBW since I have only been there a few times. But if the car keeps squirreling away capacity and there is some to be had down there, I may eventually start using that. I appreciate the reports of capacity that Ed makes for driving below VLBW because it gives a glimpse of what may be going on with the LBC.

Same here: degradation (I'm very close to dropping my 4th bar) has required me to run below VLBW far more than I ever did...and I'm going farther than I ever thought possible (thanks to LeafDD). The GIDs are definitely getting "bigger"! However, I can't say I enjoy playing Russian Roulette with turtle mode (which I finally hit for the first time after 4 years of ownership).

 

[edatoakrun]

I don't think you folks are likely to get very far if you continue to try to interpret kWh use by starting with gibberish from your LBCs.

To repeat, I have been tracking my available battery capacity for over four years, both the energy use side, through range tests normalized by time and both ambient and battery temperature, and recharge capacity, without allowing the LBC data to contaminate the process.

I'd suggest you all try to do the same, and I'd welcome any substantive criticism of my own methods.

drees :
...So let's look at my most recent LBW to 100% charge data... I probably lost...a total of 23% loss of capacity...

Are you are positing that the LBC capacity loss report is intended to reflect only that fraction of the available kWh, from a 100% charge to LBW?

Well, maybe so, and maybe Nissan engineers had a reason they wanted less of the bottom of the pack utilized as the battery ages.

If using my pack below the rising LBW is problematic for battery life, I will certainly find out in the next few years, as I will need to use that charge, more and more.

FYI, it looks like my loss of available actual kWh from "100%" to LBW averages ~21%, from , 12.9 kWh/16.3 kWh =~79%., on my usual test route.

Since the LBC does not represent a fixed percentage of capacity, but varies over different routes with different ascent decent profiles(and probably also with battery temperature) I'd probably get a different average if I regularly tested my pack on a different route.

My LBC now reports ~75%.

I believe the % of my pack's capacity from LBW to turtle (never gone past that) has probably increased by ~11.5%, from ~4.7 kWh when delivered, to ~5.2 kWh presently.

So, I believe my total average available capacity from "100%" charge to turtle has decreased from ~21 kWh to ~18.1 kWh, or between 13% and 14% since delivery, following 16 Amp ~240 V charges, ending at ~80 F battery temperatures.

Again, this is calculated on my regular test route, though it appears I would have been able to use ~18.5 kWh if I had driven past Corning and all the way to turtle last week, with a ~25 F warmer battery, and ~7,000 ft. less ascent and descent, than on my usual test route.

drees:
...As an end user all we have to go off are what the car tells us. And if the car tells us we have lost 23% between 100% and LBW - why should we have to assume otherwise? Nissan had multiple chances to fix the car to do the right thing...

IMO Nissan has treated LEAF owners with something just short of contempt, as if we are all childish simpletons who will believe whatever the capacity bars or LBC data show, as interpreted only by the bizarrely limited explanation in the capacity warrantee, that four bars lost represents ~30% of capacity loss.

Of what capacity?

Total?

Or if "30%" is of available capacity, from "100%" to where...LBW, VLBW, turtle or dead?

But (assuming Nissan is no more forthcoming in the future) there's no reason we have to act like helpless victims, curtailing our driving just because the LBWs and VLBWs come earlier during trips.

I intend to continue driving my LEAF on the OE battery until I find an alternative pack, or another BEV, that works better for me.

And I really don't give a damn when I get the LBWs and VLBWs, as long as I know that My LEAF has the range I need for my trip.

 

[TomT]

A VERY big Plus One!

IMO Nissan has treated LEAF owners with something just short of contempt, as if we are all childish simpletons who will believe whatever the capacity bars or LBC data show, as interpreted only by the bizarrely limited explanation in the capacity warrantee, that four bars lost represents ~30% of capacity loss.

 

[RegGuheert]

I don't think you folks are likely to get very far if you continue to try to interpret kWh use by starting with gibberish from your LBCs.

It's not all gibberish. Ingineer has tested the system and claims that the voltages reported are *extremely* accurate. At the end of the day, it is those voltages, and ONLY those voltages which dictate the operation of the shunts and determine when the charger stops charging and when the battery disconnects.

What would be interesting would be a detailed chart of the battery voltages at idle all the way down to disconnect. I have compiled a set of that data, but I can only correlate it to the *reported* SOC and my data only goes down to VLBW or slightly below. Does anyone know the exact voltage of the lowest cell when the battery disconnects?

FYI, it looks like my loss of available actual kWh from "100%" averages ~21%, from , 12.9 kWh/16.3 kWh =~79%., on my usual test route.

So my total average available capacity has decreased from ~21 kWh to ~18.1 kWh, or between 13% and 14% since delivery.

Again, this is calculated on my regular test route,...

Those two statements appear to be in direct conflict unless I assume the first one is from 100% to LBW while the second is from 100% to turtle (or disconnect). Is that what you meant?