Discuss data from the LEAF Battery app, and Comparisons

[mwalsh]

6/22: AHr=54.07 CAP=81.61%
6/24: AHr=54.01 CAP=81.52%
6/25: AHr=54.12 CAP=81.68%
7/11: AHr=53.16 CAP=80.23%
7/12: AHr=53.30 CAP=80.45% Hlth=76.57%
8/26: AHr=52.54 CAP=79.30% Hlth=75.16%
9/11: AHr=51.49 CAP=78.49% Hlth=73.22%
10/2: AHr=51.81 CAP=78.98% Hlth=73.82%
10/8: AHr=51.58 CAP=78.63% Hlth=73.39%

10/23: AHr=52.12 CAP=79.51% Hlth=74.46%

Values on the rise again. Could be the car finally agrees that summer's over.

 

[TonyWilliams]

Values on the rise again. Could be the car finally agrees that summer's over.

If you're on the old firmware, probably.

 

[Weatherman]

Still spiraling into the abyss, here:

July 11th: 57.45 Ahr
Aug 12th: 56.30 Ahr
Sept 10th: 55.46 Ahr
Today: 54.61 Ahr

Battery temp at midnight, last night, after yesterday's L2 charge completed, was between 91F and 94F.

 

[mwalsh]

Battery temp at midnight, last night, after yesterday's L2 charge completed, was between 91F and 94F.

AM battery temps here, after resting 10 hours before a L2 charge of 4 hours, were just 72.5-75.2 degrees. Is there a reason you charge in the evening, instead of leaving time for the pack to cool some?

 

[Weatherman]

Last summer I charged early in the morning, but then the battery was very warm to start the day and got hotter as the day progressed. By charging in the evening, the battery at least has a chance to cool off into the upper 80s by the time I leave for work in the morning instead of staying in the 90s all the time (the car sits outside all night).

But, it's, really, six of one and half dozen of another.

One thing about living in a tropical environment, close to a warm ocean, is that there is very little difference between the air temperature at night and the air temperature during the day. Take a look, here, and you'll see what I mean:

http://www.srh.noaa.gov/mfl/?n=cliplot" onclick="window.open(this.href);return false;

 

[RegGuheert]

Battery temp at midnight, last night, after yesterday's L2 charge completed, was between 91F and 94F.

AM battery temps here, after resting 10 hours before a L2 charge of 4 hours, were just 72.5-75.2 degrees. Is there a reason you charge in the evening, instead of leaving time for the pack to cool some?

Battery temps just after L2 charging from 33% to 94% at 3:00 PM today were 50.3F to 51.6F. (First frost is expected this evening.)

I finally got my ELM327 this week, so I can start reporting capacity numbers here (no P3227 update):

09/28: Odo=13,877 AHr=59.18 CAP=90.21% Hlth=87.89%
10/24: Odo=14,515 AHr=59.39 CAP=90.53% Hlth=88.31%

 

[Weatherman]

Fortunately, our hot, rainy season came to an end last night (about five days later than normal). We'll be in the warm, dry season for the next six to seven months.

The battery temp fell from the low 90s last night to within a couple of degrees either side of 80F, by early this morning (air temps were in the mid 70s, overnight), and should remain mostly in the 70s and 80s until next spring. Should slow down the degradation rate by a tiny bit.


It has been noted by others that "spirited" driving of the car warms the battery as it discharges. I didn't keep close track of this over the summer, but made it a point to make a note of it this morning. Left for work with the battery temp around 80F. After a 20 minute drive at an average speed of around 65, the battery temp warmed by one degree to 81F. Air temps were in the mid 70s. Unless it take a few hours for the temperature sensors to register the rise in temp, I just don't see it. (I guess I'm very fortunate I don't see it. It was very common, over the summer, to start the day with battery temps in the lower 90s. I can only imagine the sad shape my LEAF's battery would have been in if the morning drive routinely pushed the battery temp over 100F.)

 

[RegGuheert]

It has been noted by others that "spirited" driving of the car warms the battery as it discharges. I didn't keep close track of this over the summer, but made it a point to make a note of it this morning. Left for work with the battery temp around 80F. After a 20 minute drive at an average speed of around 65, the battery temp warmed by one degree to 81F. Air temps were in the mid 70s. Unless it take a few hours for the temperature sensors to register the rise in temp, I just don't see it. (I guess I'm very fortunate I don't see it. It was very common, over the summer, to start the day with battery temps in the lower 90s. I can only imagine the sad shape my LEAF's battery would have been in if the morning drive routinely pushed the battery temp over 100F.)

Actually, if your battery temperature starts above 90F, I suspect that the part of your battery which matters most does rise above 100F during your 20-minute commute. Here's why: While the battery sensors measure the temperature on the case of some of the modules, the heat is dissipated inside the chemistry in the four pouch cells found within the modules. The four cells are stacked so the two in the middle of the stack will have the highest temperature rise, since they have the highest thermal resistance to ambient.

Here is a simplified thermal schematic of the situation using (rather poor) ASCII art:

                    Outer Cell     Sensor    Air Below Pack
Inner Cell_____/\/\/\_______/\/\/\_______/\/\/\_______/\/\/\____ Ambient Air
           |    RCell   |    Rcase   |    Rair    |  Rambient
          ___          ___          ___          ___
          ___          ___          ___          ___
           |            |            |            |

- When the LEAF is sitting, the thermal resistance Rair (from the bottom of the pack to the air between the pack and the bottom panel) increases because that air is no longer flowing over the pack. Also, because of the bottom panel, this air becomes "dead air" and insulates the pack from ambient. This panel gives rise to an additional thermal resistance when sitting that I have labeled Rambient.
- If the car sits for a long time without charging in a constant-temperature environment, all five nodes in this diagram will eventually be at the same temperature. That means that what the temperature sensors report will match both the ambient air temperature and the Inner and Outer Cell temperatures.
- If the car is charging while sitting, heat will be dissipated into the nodes labeled Inner Cell and Outer Cell. This will cause all the temperatures to rise except Ambient Air, with the highest temperature at Inner Cell and gradually decreasing from left to right in the diagram. During a Quick Charge session from empty to full, I expect tha temperature inside the Inner Cell could exceed the Sensor temperature by more than 20F and the Ambient Air temperature by as more than 40F. (Someone can easily check me on part of this: If you perform a QC from empty to full, do you see the temperature reported by the sensors rise about 20F above ambient? If not, how much do you see?)
- When the vehicle is moving, the thermal characteristics change fairly significantly. In that case, the Air Below Pack is no longer "dead air", but rather it now is moving ambient air. This change has two effects: First, it essentially eliminates Rambient since the temperature on both sides of that bottom panel is the Ambient Air temperature. Second, it significantly reduces the thermal resistance of Rair, the resistance from the bottom of the pack to the air below it. In fact, this is the main thermal design feature of the LEAF. As a result of this, you can see that while you drive the LEAF, the thermal resistance between the temperature sensors and the Ambient Air temperature is very low. It is made up of only the resistance of the steel of the battery modules and bottom plate and the resistance from the bottom plate to the moving air below. For this reason, you will typically not see a large temperature difference between these sensors and the ambient air while driving unless the entire battery pack is already at a very high temperature, such as it might be following multiple QC sessions.
- Also when you are accelerating or decelerating, heat is being dumped into the nodes labeled Inner Cell and Outer Cell, which represent the insides of the pouch cells in the battery modules. This causes those nodes to heat up to temperatures which are above that reported by the battery sensors. If you drive at high speeds for a significant time, I expect the temperatures inside the Inner Cell could get 10F or more higher than the temperature reported by the sensors.

For reference, here is a discussion from Phil http://www.mynissanleaf.com/viewtopic.php?f=30&t=8802&p=206096&hilit=+temperature+rise#p206096 indicating what he sees at the sensors after driving and letting the car sit for a while:

I've noticed I can start out with a pack temp of 70, then take a long drive and return home with it reading only a few degrees higher, but still slowly climbing. It will continue to rise once the car is parked (even when inside a building with cool temps and not charging) and sometimes I've seen it get close to 80 before it starts to cool back off. This process takes hours because of the large thermal mass of the pack. If you start charging right away it exacerbates this issue and the temp climbs even higher. Another argument for charging early in the morning!

Note that if the sensors come up to a temperature of 80F after sitting for some hours, the temperature inside the Inner Cells would have to be significantly above 80F at the end of the drive for this to happen. Keep on mind that during those hours of sitting, the various temperatures within the battery are not only equalizing, but some of the heat is lost to ambient during that period.

Based on all of the above, it may turn out that in a very hot climate, charging just befor departure might not give the lowest peak battery temperatures. Perhaps charging from around midnight to 5AM would work out to be best in that case.

 

[Weatherman]

I suspected there might be a delay between the time internal heat is generated and the time the temperature sensors show it.

Yesterday evening, I checked the battery temp when I left work. It was 80F. When I arrived home, it was at 81F. Three hours later, when I started the evening charge cycle, it was 84F. Air temp, during that time, was in the upper 70s. So, unless the temp peaks at a higher level during the three-hour interval between stopping the car and starting the charge, I got a 3F rise. The temp was 87F when the charge was complete, three hours later. I don't know if the temp continues to rise after the charge cycle, but if it does, at least it occurs in the wee hours of the morning when air temps are at their coolest. Temp was back down at 80F, when I left for work this morning.

All of this is, pretty much, academic, anyway. Even though I was charging the car during the early morning hours, the first year I owned it, it still lost the first capacity bar before the year was out. The rate of decay is not significantly worse or better since I changed my routine.

 

[mxp]

Posted first Jun 28, 2013:

2011 Leaf SL (July 1, 2011 pickup)
3 days a week 100% charge (overnight); other days 80%.
24k miles ODO


Warm morning: June 28, 2013. 6 temp bars showing even before drive.
Data below taken after 100% overnight charge, about 8am:
80def F
AHr=58.84 CAP=88.81%
91.5% SOC 242 gids

Cold morning: Oct 25, 2013. 5 temp bars showing before driving.
Data below taken after 100% overnight charge, about 8am:
63def F
AHr=57.87 CAP=87.87%, Hlth= 85.33%
92.4% SOC 240 gids

Assuming if I lose 1% CAP every quarter, I suspect I will be on track to losing my 1st Bar by end of next year (4 quarters) as the CAP drops down to ~85%.

Is that about right?

 

[RegGuheert]

Reported capacity is creeping up (pre-P3227)

09/28: Odo=13,877 AHr=59.18 CAP=90.21% Hlth=87.89%
10/24: Odo=14,515 AHr=59.39 CAP=90.53% Hlth=88.31%
10/25: Odo=14,595 AHr=59.47 CAP=90.66% Hlth=88.46%

This is the first time we have had the LEAF below LBW since getting LEAFSpy, so I was interested to find out what the battery characteristics looked like there. Here are the numbers:

SOC - 17.9%
Highest Cell - (Cells 65 and 66) - 3.742V
Lowest Cell - (Cell 26) - 3.727V
Average Cell - 3.735V
Cell Range - 15mV
Temps: 53.0F 53.0F 50.3F 53.0F

Note that the voltage difference is low at only 15 mV. Given the battery was fully charged yesterday and balanced at the top end with the following characteristics:

SOC - 94.0%
Highest Cell - 4.106V
Lowest Cell - 4.090V
Average Cell - 4.100V
Cell Range - 16mV
Temps: 51.6F 51.6F 50.3F 51.6F

So, it appears the pack is balanced and the cells are pretty well matched since they track over most of the range of charge.

Interestingly, the temperature when my wife left this morning was about 32F (the LEAF sat outside the garage all night) and it was only 45F when she returned, yet most of the temperature sensors read 53.0F when she arrived home. Note that she was driving on low-speed roads with speed limits of 25, 35, 45, 50 and 55 MPH (only about 1 mile at 55). She was hypermiling at the end of the drive, but the heater was on, also. An 8F temperature rise seems pretty high for this situation, particularly since the ambient temperature was rising at the time, but I will admit that it is somewhat welcome in the cold weather!

 

[DaveEV]

So, it appears the pack is balanced and the cells are pretty well matched since they track over most of the range of charge.

Yep, pack seems very well balanced. How often do you charge to 100%?

Interestingly, the temperature when my wife left this morning was about 32F (the LEAF sat outside the garage all night) and it was only 45F when she returned, yet most of the temperature sensors read 53.0F when she arrived home. An 8F temperature rise seems pretty high for this situation, particularly since the ambient temperature was rising at the time, but I will admit that it is somewhat welcome in the cold weather!

Very interesting. My hypothesis is that the internal resistance of the pack goes up in cold weather so it heats up more.

This would also help explain why Weatherman sees less heating than others - his pack starts out significantly warmer.

 

[RegGuheert]

I suspected there might be a delay between the time internal heat is generated and the time the temperature sensors show it.

Agreed. The funny-looking things at the bottom of my ASCII-art diagram represent the thermal capacitance which accounts for the delay.

Yesterday evening, I checked the battery temp when I left work. It was 80F. When I arrived home, it was at 81F. Three hours later, when I started the evening charge cycle, it was 84F. Air temp, during that time, was in the upper 70s. So, unless the temp peaks at a higher level during the three-hour interval between stopping the car and starting the charge, I got a 3F rise.

Actually, 84F indicates the sensors were 6F or 7F above ambient ("high 70s"). That means that the Inner Cell temperature must have been quite a bit above that point at the time you turned off the car. Perhaps 90F or more.

The temp was 87F when the charge was complete, three hours later. I don't know if the temp continues to rise after the charge cycle, but if it does, at least it occurs in the wee hours of the morning when air temps are at their coolest.

The temperature at the sensors will continue to rise after the completion of the charge cycle, but the key temperature that affects degradation rate, which is inside the cells, begins to drop immediately after completion of charging.

Temp was back down at 80F, when I left for work this morning.

Do you know the outside temperature at that time?

All of this is, pretty much, academic, anyway. Even though I was charging the car during the early morning hours, the first year I owned it, it still lost the first capacity bar before the year was out. The rate of decay is not significantly worse or better since I changed my routine.

Agreed. It is quite academic. The main point I want to make is that the temperature of the sensors is NOT the temperature that we care about for determining battery degradation and the one we care about certainly peaks at a higher value than the sensors do. Also, the temperature reported the sensors is closer to the ambient air temperature than the Inner Cell temperature while driving but is farther from the ambient air temperature when charging due to the difference in thermal resistances that exist in those two cases.

 

[RegGuheert]

So, it appears the pack is balanced and the cells are pretty well matched since they track over most of the range of charge.

Yep, pack seems very well balanced. How often do you charge to 100%?

At least once per week. But usually the LEAF sits at 20% to 50% SOC and we only charge it up enough to return back there (typically 6 to 9 bars).

Interestingly, the temperature when my wife left this morning was about 32F (the LEAF sat outside the garage all night) and it was only 45F when she returned, yet most of the temperature sensors read 53.0F when she arrived home. An 8F temperature rise seems pretty high for this situation, particularly since the ambient temperature was rising at the time, but I will admit that it is somewhat welcome in the cold weather!

Very interesting. My hypothesis is that the internal resistance of the pack goes up in cold weather so it heats up more.

This would also help explain why Weatherman sees less heating than others - his pack starts out significantly warmer.

If you are saying that the electrical resistance goes up in the cold, meaning there is more power dissipated, I can certainly agree with that. That makes good sense!

If you are saying that the thermal resistance goes up in the cold, then I'm not sure how that would happen. If that is what you mean, can you please explain further?

 

[DaveEV]

If you are saying that the electrical resistance goes up in the cold, meaning there is more power dissipated, I can certainly agree with that. That makes good sense!

Yep, that's what I'm saying. That's why Nissan mentioned that driving the car in really cold weather was plenty to keep the battery "warm" and that pack heating is only needed if the car is sitting in extreme cold for extended periods.

 

[Weatherman]

Temp was back down at 80F, when I left for work this morning.

Do you know the outside temperature at that time?

It was 75F.

The main point I want to make is that the temperature of the sensors is NOT the temperature that we care about for determining battery degradation and the one we care about certainly peaks at a higher value than the sensors do. Also, the temperature reported the sensors is closer to the ambient air temperature than the Inner Cell temperature while driving but is farther from the ambient air temperature when charging due to the difference in thermal resistances that exist in those two cases.

Which makes the lack of an active cooling system even worse than first thought.

We have heat from charging and discharging, which doesn't even show up at the car's temperature sensors for hours after it occurs. So, parts of the battery could be several degrees higher than the battery temps sensors show.

And, we have a slow heat dissipation rate, so the average hourly temperature at the sensors can, easily, be around 10F warmer than the average air temp (meaning if the average, hourly air temp is in the lower 80s, the average hourly battery sensor temp is in the lower 90s).

 

[GregH]

My capacity has been inching up in the cooler weather as well (much more 5TB vs constant 6TB).. ie low-mid 20s vs high 20s to mid 30s.
Oct 7: 58.8628Ah 86.27hx (the low point)
Oct 25: 60.1945Ah 89.89hx this morning

The odd observation is that the Ah/Hx keep inching up (pre P3227 btw) but this morning only charged "full" to 249Gids, 92.7%SOC which is a tad on the low side for me.. would have expected more like 255 Gids and 95%. Perhaps discrepancies in calculated vs true capacity are more likely to show up as a less than full charge rather than a snap change down at VLBW.. This was a full charge from just under LBW. It'll be interesting to see if my next full charge goes to a higher SOC/Gid count and if that knocks my Ah/Hx numbers back down a tad.

 

[Stoaty]

It seems odd to me that people are discussing the vagaries of firmware known to give an inaccurate reading of the Leaf battery capacity. Why not get the P3227 update (includes compatibility fixes for some EVSE) and discuss the more accurate readings?

 

[Weatherman]

It seems odd to me that people are discussing the vagaries of firmware known to give an inaccurate reading of the Leaf battery capacity. Why not get the P3227 update (includes compatibility fixes for some EVSE) and discuss the more accurate readings?

Makes people feel good seeing their battery capacity go up.

"Hope, it is the quintessential human delusion, simultaneously the source of your greatest strength, and your greatest weakness."

 

[surfingslovak]

The main point I want to make is that the temperature of the sensors is NOT the temperature that we care about for determining battery degradation and the one we care about certainly peaks at a higher value than the sensors do. Also, the temperature reported the sensors is closer to the ambient air temperature than the Inner Cell temperature while driving but is farther from the ambient air temperature when charging due to the difference in thermal resistances that exist in those two cases.

With all due respect, I don't think that this matters very much. The car will spend the majority of its life just sitting, not charging and not driving. While what you mentioned is good to keep in mind and to consider, I don't believe that it will turn out to be very significant. I would also point out that we had another rather academic debate about the error of the voltage sensor last year. While these discussions can be entertaining, educational and perhaps even illuminating, it would be good if they did not shift away focus away from more significant factors and observations. There were numerous theories that did not translate to any observable impact on battery life, such as increased solar loading due to the color of the car. Perhaps even the difference between parking in an enclosed garage versus an open driveway or a carport won't result in major observable differences. This does not mean that these points are not correct or valid. As always, time will tell.