Leaf Miles / KWh is Wrong Or Usable bat. cap. is not 24 KWh

[Herm]

Ofcourse Nissan knows exactly how much they use - they designed this. They don't want to talk about it. When I chatted with Mark Perry, he deftly dodged the question & I didn't press.

I wonder whey they are so coy about it?, perhaps so that spec war does not get started.. "GM announces a BEV with a 25kwh pack"

Other manufacturers appear to be more open with the info.

 

[DaveinOlyWA]

I think part of it is avoiding. A pudding match over performance verses economy discussion. At 3400 lbs. We wont win a lot of those battles

 

[evnow]

Other manufacturers appear to be more open with the info.

Not sure about that. We have been told Focus has 23 kwh. Is that total or usable ?

GM danced around usable for a long time (started with 8 kwh and ended with 10+).

 

[planet4ever]

The in-car reported miles/KWh over a long test period is 3.8 miles/kwh. Average distance over a wide variety of trips over the same period, I seem to be averaging 5.5-6 miles/bar. This works out to ~1.5KWh/bar.

The first bar seems to disappear more quickly than the others (whether you are charging to 80% or 100%). Since you run through that first bar for every charging cycle, and through other bars less frequently, this gives a false downward bias to an "average number of bars" measurement. In my own limited experience if your miles/kWh is 3.8 you should be getting at least 7 miles/bar, excluding the first one.

That would change your calculation to ~1.8 or even ~1.9 kWh/bar. Taking into account the "smaller" first bar, let's say we have 11.5 bars, and multiplying gives -- guess what -- just over 21 kWh. But with the updated firmware there seems to be a significant reserve after the last bar is gone, probably at least as much as a bar, maybe a bar and a half. This brings us up to 23 or 24 kWh available.

Charging times roughly correspond to this number. About 30 min/bar each night. (at 3.3KW from the wall)

Ah, but have you actually measured 3.3 kW as your charging rate at the wall? The charger itself is rated at 3.3 kW, but I think that is its output rating, not its input rating. After all, 240v * 16A = 3.84 kW. So if you were figuring 3.3/2 = 1.65 and rounding down to 1.5 to allow for charging loss, you may have been making a bad assumption. Also, your 30 min/bar seems low. The owners manual say ~7 hours from low battery warning to full. Low battery warning now seems to come in about the middle of the last bar. Assuming half a bar there, and half a bar at the top, we're talking 7 hours for 11 bars, or about 38 min/bar. That, with 3.3 kW going into the battery, would give ~2.1 kWh/bar.

Now, I suspect my numbers are high, and yours are low, but I think my assumptions are about as reasonable as yours.

Ray

 

[awolfe63]

That's my point. I am recharging at 30 min/bar. That is too fast if the battery is 24WH.

In any case, I'm not trying to prove anything about my battery. I'm trying to figure out how I can get Nissan to test my battery and tell me whether it is in spec or out of spec.

 

[Ingineer]

Not too long ago, I was able to spend several hours with a LEAF and Consult III+ (The official Nissan scantool).

Of the many thousands of parameters that can be queried, one of them is the total battery capacity in Amp-Hours. The LEAF I tested reported 67.568Ah. Note that battery pack capacity is never measured or expressed in kWh, rather Ah. If you know the discharge curve voltages, you can calculate kWh however.

The Battery ECU keeps track of this Ah figure and adjusts it dynamically. It's used to calculate the actual SOC, so this is the actual real capacity in Amp-Hours. How much of this is "usable" in the car is another matter, you have to know what metrics the car uses to base it's cut-off point at.

-Phil

 

[Herm]

Of the many thousands of parameters that can be queried, one of them is the total battery capacity in Amp-Hours. The LEAF I tested reported 67.568Ah. Note that battery pack capacity is never measured or expressed in kWh, rather Ah. If you know the discharge curve voltages, you can calculate kWh however.

If they charge the batteries to 4.15V per cells then that is 27kwh, but they may not charge to that high a voltage.. is voltage displayed by the scan tool?..

 

[Ingineer]

If they charge the batteries to 4.15V per cells then that is 27kwh, but they may not charge to that high a voltage.. is voltage displayed by the scan tool?..

Yes, voltage and amps are available. Problem is you can't go by the maximum voltage, you have to go by the voltage in real-time. In other words, If you are drawing 50 amps from the pack and it's at 400v, that would be 20kW of load, if you sustain that exact load for 1 hour, then that's 20kWh. Unfortunately, the battery voltage is not constant, so even if the amps stay constant, the total kW will fall as the cell voltages fall. You'd also have to account for the voltage drop in the cell under load if you want to be accurate.

This is why battery capacity is measured in amp hours.

You can "back of the napkin" it and go with whatever the voltage is at about 50% SOC and get reasonably close though.

-Phil

 

[Smidge204]

Each cell is rated by the manufacturer at 3.8V, 33.1Ah (0.3C)

96 cells in series gives 364.8 volts nominal pack voltage. 364.8V * 33.1Ah = 12.1kWh per series stack of 96 cells, times two stacks in parallel for 24.2kWh total.

So 24kWh is absolutely a number related to the battery pack. Of this there should be no dispute! :lol: The contention is if this is the total absolute capacity or the customer-accessible capacity.

To try and resolve this I refer to the EPA testing procedure and data. Testing procedure is summarized as follows:

Electric Vehicle - City Test Procedure Summary - Following SAE J1634 Recommended Practice, the battery is fully charged, the vehicle is parked over night, and then the following day the vehicle driven over successive city cycles until the battery becomes discharged (and the vehicle can no longer follow the city driving cycle). After running the successive city cycles, the battery is recharged from a normal AC source and the energy consumption of the vehicle is determined (in kW-hr/mile or kW-hr/100 miles) by dividing the kilowatt-hours of energy to recharge the battery by the miles traveled by the vehicle. ... The city driving range is determined from the number of miles driven over the city cycle until the vehicle can no longer keep up with the driving cycle.

From the data we see the EPA arrived at 34 kWh/100 miles (combined city/hwy) and 73 miles or range. (Note: These figures are after their driving model adjustments - which essentially reduce everything the actually measure by by 30%). I would use the unadjusted figures but they are not complete and I want to interpolate as little as possible. So:

34/100*73 = 24.82 kWh.

Note that the EPA did not drive the car until it was dead, but until the car could no longer maintain prescribed cycle performance. So the battery was not completely drained of it's user-accessible capacity. It is not stated anywhere in the documentation but I suspect the test ended when the vehicle entered turtle mode, or shortly thereafter, and the vehicle could no longer maintain acceleration or top speed required by the test. If we include charging efficiency of ~90% we reduce the actual to-battery recharge to 22.34kWh, but since the battery was not likely fully drained the 1.7kWh shortfall could easily have still been in there.

In conclusion, based on EPA testing, I propose that the total user-accessible capacity of the battery pack is approximately 24kWh and that the manufacturer's ratings on the cells include the "safety factor" for prolonged cell life. The only other alternative is that the EPA somehow managed to bypass the limiting tot use the full as-installed 100% of the pack which is highly unlikely.

I further propose that the car's internal telemetry is faulty (as indicated by numerous anecdotes on this forum alone) and so any method to estimate the battery capacity using them is suspect.
=Smidge=

 

[evnow]

In conclusion, based on EPA testing, I propose that the total user-accessible capacity of the battery pack is approximately 24kWh and that the manufacturer's ratings on the cells include the "safety factor" for prolonged cell life. The only other alternative is that the EPA somehow managed to bypass the limiting tot use the full as-installed 100% of the pack which is highly unlikely.

EPA doesn't give the m/kwh that Leaf showed at the end. But we have it now and that shows Leaf doesn't let you use the full 24 kwh. That is what this thread is talking about.

 

[Herm]

Yes, voltage and amps are available. Problem is you can't go by the maximum voltage, you have to go by the voltage in real-time.

Yeah, had a brain fart.. you have to use the average voltage to calculate the capacity.. what pack voltage does the scan tool show when the battery is charged to 100%?, 80%?.. that would tells us what Nissan is leaving as a margin on the upper end.

 

[Smidge204]

EPA doesn't give the m/kwh that Leaf showed at the end. But we have it now and that shows Leaf doesn't let you use the full 24 kwh. That is what this thread is talking about.

Yes. And my point, again, is the display in the car is probably wrong. At the very least the anecdotes on this forum give reason to consider the possibility. You always seem to ignore that part of my argument. :? Why should we care what the car says when it is the car we suspect of being wrong?

Specifically, I believe the mi/kWh displayed on the dash and reported by Carwings is higher than actual, possibly by as much as 20-30% in some cases.

This hypothesis is backed up by independent testing which does not rely on the figures both by the government and individuals using from-the-wall power readings. It is consistent with other people's observations and it makes sense in light of other issues such as the remaining range indicator being damn near useless.

You even acknowledge this discrepency several times, calling Carwing's reported mi/kWh "optimistic." I contend that the dash displayed value is similarly skewed, and you admit this possibility as well in your very first post of this thread.

Not sure what to tell you beyond that. All your evidence that the usable capacity is less than 24kWh relies on the car's systems. There is evidence the car's systems are not accurate. There is evidence that does NOT use the car's systems which indicates the capacity is a full 24kWh.

Edit: Heck, I just re-read this entire thread to be sure I didn't owe anyone an apology, and you even agree with me that the mi/kWh reported by the car is bloated - but then you use the admittedly bloated figures to calculate that the battery capacity is less than advertised. Am I the only one who sees the problem with this?
=Smidge=

 

[bowthom]

Hello,
I've been closely tracking carwings and blink since March and I agree with both of you. My car is exhibiting both scenarios at different times. Sometimes the SOC falls fast and it only takes 60% of the indicated consumption to charge it back up. Other times carwings reports 40% less power usage than it actually takes. Unfortunately, the blink network has reporting errors too and some days I don't get any good data. The empirical part of the observation involves time. For example, I ran the battery down to 1 bar and it only took three hours to charge up. Not possible even if the battery capacity was only 20 kWh. I'm going to keep track of it very closely so I have lots of data to back up my concerns. I don't want look like I'm crazy at my 3500 mile checkup. :lol:

 

[DaveinOlyWA]

now that we have gotten absolutely NO WHERE with what is wrong, lets talk about what is right.

the flash update had greatly improved the estimated range to the point, that i took off with a full charge the other day, it showed me 112 miles of range. i drove it 85 miles and parked it and it showed me having 24 miles left.

now, granted, this was all surburban low speed driving with a bit of traffic congestion for probably 50% of it. so a bit of pluses and minuses to the range delivered.

all in all, it looks like the car can easily do 100 miles if driven reasonably. i avoid freeways (that has nothing to do with driving a Leaf) since for the most part, they are not convenient to my destinations unless going out of town.

but i still feel that i have not reached the best part of the year for maximum range. so we shall see.

i no longer care how much battery capacity i have to use

 

[evnow]

Yes. And my point, again, is the display in the car is probably wrong. At the very least the anecdotes on this forum give reason to consider the possibility. You always seem to ignore that part of my argument. Why should we care what the car says when it is the car we suspect of being wrong?

Specifically, I believe the mi/kWh displayed on the dash and reported by Carwings is higher than actual, possibly by as much as 20-30% in some cases.

I've explained in the OP that m/kwh could be wrong. Also let us keep CarWings out of this as it is acknowledged to be wrong by Nissan CS.

Since I started the thread, the balance of evidence is shifting towards usable capacity being less than 24 kwh. Let us start with your earlier post about EPA findings (see more detailed calculations I had posted earlier in your foia thread http://www.mynissanleaf.com/viewtopic.php?p=54449#p54449).

34/100*73 = 24.82 kWh.

Note that the EPA did not drive the car until it was dead, but until the car could no longer maintain prescribed cycle performance. So the battery was not completely drained of it's user-accessible capacity. It is not stated anywhere in the documentation but I suspect the test ended when the vehicle entered turtle mode, or shortly thereafter, and the vehicle could no longer maintain acceleration or top speed required by the test. If we include charging efficiency of ~90% we reduce the actual to-battery recharge to 22.34kWh, but since the battery was not likely fully drained the 1.7kWh shortfall could easily have still been in there.

The problem with the above starts with ~90% figure. Let us, instead, use the 85% industry standard,

24.82*85% =21.1 kwh.

We know that after turtle, even at low speeds, Leaf doesn't go far (in the old firmware). If that was 1.7kWh, you'd expect to get some 8 miles at low speeds. People have reported 3 or 4 miles max (IIRC) and down to even a few hundred feet. Edmunds reported 2 miles in their 132 mile test. So, let us call it a generous 1 kWh - bringing us to 22 kWh.

Also, from my earlier post, in this thread ...

Notice that none of this uses Nissan's shown m/kwh.

So, in conclusion, I conclude that the below conclusion is not likely

In conclusion, based on EPA testing, I propose that the total user-accessible capacity of the battery pack is approximately 24kWh...

Note that if the usable is not 24 kwh, it doesn't mean the total is 24 kwh. It can be higher at 27 kwh, as reported in the thread.

Now, if we take the usable to be 21 to 22kwh, m/kwh starts making good sense and agrees with various numbers that have been reported. So, I no longer see any reason to assume that Leaf's dash/console m/kwh is wrong.

 

[Smidge204]

We do not know if the car entered turtle mode during testing. The test ends when the vehicle can no longer perform the test's requirements. Turtle mode seems a reasonable reason for this to happen, but it could have happened sooner. Does the LEAF reduce power/top speed before entering turtle mode?

And regarding the older FOIA thread: Even you got ~24kWh with two different sets of data "wall to wheels" - which would thus include charger losses.

While I understand what you're saying, the reason I'm sticking with vehicle error is as follows:

In favor of a 24kWh battery pack:
-EPA test data and filing paperwork
-Battery manufacturer's data
-Nissan's claims
-Some owner's reports

In favor of <24kWh:
-Some owner's reports


It is known the car has problems when it comes to power management. It could be poor software engineering. It could be a last minute change in specs for the power measuring hardware, or a calibration error. In fact I'd favor a calibration error since the magnitude of the problem seems to vary. But I still think it is somewhat more likely that the mi/kWh as reported by the car is in error than there being some kind of conspiracy perpetrated by Nissan and a professional laboratory operated by the federal government.

Of course the only way to be sure would be to install our own power metering equipment and see what's up. God only knows when the LEAF will even be available in my area so I have to rely on someone else to get their hands dirty!
=Smidge=

 

[awolfe63]

Some more data. Ran my car well past 1 bar Saturday. Remaining range was reported at 7 mi. (but low charge light had just gone on and no turtle mode) Charging to 80% required 4:00 even. That means that I added at most 13.2KWh to the pack to get back to 80% charge. If I had 10% or usable charge left (generous), then that works out to 18.8KWh usable. This still seems way off of what other people are seeing and certainly much less than 24KWh.

I got 53 miles out of that day at a reported 4.1M/KWh. That works out to 12.9KWh used. Pretty close to the charge I calculated.

I'm going to keep pushing my dealer to do a diagnostic. Whatever "spec" is, I don't think my car is measuring up.

 

[planet4ever]

First there is low battery, and then very low battery, and then turtle. Assuming you have the upgrade, VLB and turtle both happen after the last bar disappears. So if you still had a bar showing you were a long ways from empty. Just my personal opinion, but I think you still have about 15% usable capacity when LB appears.

Now let's look at your claim that you "added at most 13.2KWh." Basic electrical math says 4 hours * 240V * 16A = 15.36kWh at the wall. If the charger is 90% efficient (a reasonable assumption) then you added more than 13.8kWh. It is quite possible that the charger is as much as 92% efficient, which would yield 14.1kWh at the battery. (14.1kWh) / (80% - 15%) = 21.74kWh.

That's not the 24kWh that a number of us have been claiming, but it sure is better than 18.8kWh. My assumptions could be high or low, and you also need to take battery temperature at the end of the drive and the end of the charge into account. I don't think you have proved, "beyond a shadow of a doubt," that your usable capacity is significantly less than 24kWh.

Ray

 

[Coffee_Slurry]

At some point, an engineer, a CONSULT-III+, a LEAF and a dynamometer will all congregate in one place, and we will have our answer.

I'm also hoping some battery cells make it into a test lab that publishes research reports.

 

[gbarry42]

...low charge light had just gone on and no turtle mode....[charge time of 4 hr] works out to 18.8KWh usable. This still seems way off [...] Whatever "spec" is, I don't think my car is measuring up.

Had I written a message it would have sounded much like this. Except I haven't had the car long enough or taken careful enough data to be able to state this with confidence. Nonetheless, everything I'm experiencing points to this kind of a problem. Willing to blame it on how I drive, etc., but it's the time to charge back up that's suspicious. And in fact I did start a thread called "Does your charge time make sense?". I'm hoping those of us who think we have this problem will compare notes. I'm waiting for some better data before I post again.

This discussion about capacity is quite interesting, but beyond academics I would like to know if I just bought a 75-mile car, or a 40-mile one.