Joining the 80% Club

[evnow]

Assuming 90% to 10% usability range -

100% = 90% (duh)
80% = 74%
16.7% = 23% (battery low)
8% = 16.7% (Turtle)
0% = 10%

 

[AndyH]

Assuming 90% to 10% usability range -

100% = 90% (duh)
80% = 74%
16.7% = 23% (battery low)
8% = 16.7% (Turtle)
0% = 10%

I wish we had a chart or data from AESC that gave us SOC VS. voltage for these specific cells. I'm beginning to wonder if we have 10% at each end or 10% total - 5% at each end. If we have 5% at each end, then there may be some life improvement if we charge to less than 'consumer 100%'... Time to go back thu all the early interviews and re-remember.

 

[planet4ever]

8% = 16.7% (Turtle)

I think 8% is very suspect. The chart from the diagnosis manual says the last bar doesn't disappear until 4%. I wouldn't expect turtle mode until that bar disappears.

 

[planet4ever]

I wish we had a chart or data from AESC that gave us SOC VS. voltage for these specific cells. I'm beginning to wonder if we have 10% at each end or 10% total - 5% at each end. If we have 5% at each end, then there may be some life improvement if we charge to less than 'consumer 100%'... Time to go back thu all the early interviews and re-remember.

The only hint I've seen on this is that the steps on the capacity gauge are 6.25% except that first step is 15% and the last is 16.25%. Could this perhaps mean 8.75% reserved at the top and 10% at the bottom?

 

[AndyH]

I don't know. (and I don't like not knowing!)

AESC has updated their website with a picture of 'our' battery module:
http://www.eco-aesc-lb.com/en/bmodule.html

Weight: 3.8kg
Nominal Voltage: 7.5V
Capacity: 66Ah

We know that a module is 2P-2S - so that 7.5V nominal is 3.75V per cell - up from the old 3.6V number.

Doing the math for pack capacity using the nominal voltage gives us 48 modules X 2 X 66Ah X 3.75V = 23.76kWh

Some cell vendors give max capacity and voltage ranges (like Thunder Sky with their 4.3V max and 2.5V minimum) while others give the capacity and voltages for expected longer-term service. It's clear from the Service Manual that we have at least some top and bottom buffer, so it looks like AESC is giving a conservative 'user' capacity rather than cell max. Nissan says we have a 24kWh pack but the total capacity is proprietary.

The only other cell that I can find with a 3.75V nominal is made by E-Moli.
http://www.molicel.com/hq/product/DM_IBR26700A.pdf
AESC was reported to be adding some nickel to their cells so that sorta fits.

Check out the datasheet for the 26700 cell -- only 10% capacity loss after 1000 cycles from 4.2V to 2.0V at 1C charge/discharge - that's the best LiMn performance I've seen! E-Moli lists their voltage range as 0% and 100% SOC - and I don't know any cell that'll give 2000 cycles between 0 and 100% ultimate - so this is very likely an end-user voltage range and lifespan. If AESC has gotten even fairly close to this level of performance we're in fine shape - especially since Nissan's using more conservative voltages.

 

[TomT]

Everything I have read to date, and the numbers from current owners, makes me believe that the full capacity of the battery is what Nissan has always said it is, 24Kw, and there is no hidden top or bottom end. 24Kw period. I have seen no data to indicate a true capacity of 30Kw.

I still contend that charging to 100 percent is less stressful for the battery than discharging to 10 percent or less... This is true for almost every lithium battery ever made and I have a hard time believing that the Leaf battery would be any different.

But is charging to 90% less stressful than discharging to 18%? I think that is what you are really talking about. I continue to believe that the total capacity of the battery is about 30kWh with a usable charge range of 24kWh. In converting usable charge percent to total charge kWh, I contend (for a new battery at room temperature) that:

• "100% charge" is really about 27kWh
• "10% charge" is really about 5.4kWh
• "0% charge" is really about 3kWh
• The car will never let you go below about 3kWh or above about 27kWh

Note that I don't claim to know the answer to the question I asked. I only claim that it isn't "obvious" from prior lipo experience.

 

[AndyH]

Everything I have read to date, and the numbers from current owners, makes me believe that the full capacity of the battery is what Nissan has always said it is, 24Kw, and there is no hidden top or bottom end. 24Kw period.

I understand and agree completely - that is apparently EXACTLY what we have for our use as drivers of this car!

Here's the but... Behind the curtain, the engineers took a bunch of raw cells, combined them into a pack, decided the necessary power/range/etc. and attached a battery management system to the pack to make sure the cells stay alive.

This is exactly what happens for cell phones, laptops, and any other consumer device - they all have a management system inside the plastic case. There are exceptions - the RC model packs are raw cells in shrink wrap - no management. And many DIY EVs and Chinese scooters have no management installed.

The assortment on the table includes LiFePO4, LiFeYPO4, LiFeMnPO4, and LiCo/LiPo. Not a BMS on the bunch.

Give this paper a few minutes. It's from a pack management presentation that gives info on cells, protection, and balancing methods:
http://www.absoluteefficiency.com/LEAF/Building_Battery_webinar_3-4-09.pdf

LiPo/LiCo can be charged to 15V - once. All cells can be discharged to zero - and even reversed - once. But there's not a battery on the planet that will give years of service when it's used to 100% of it's raw capacity - and Nissan knows this. That's why they programmed the BMS to keep us from discharging below 3.2V per cell, a keep us from charging any cell higher than 4.2V. There IS more capacity on each end - but 'our' 100% SOC is less than - smaller than - the pack's absolute capability. It's that more conservative use that we exchange for 10 years of driving.

I have seen no data to indicate a true capacity of 30Kw.

And I haven't either. But what we do have points strongly in that direction. Nissan stated that there is about 10% capacity above and below the user capacity zone. In another Q/A Nissan was asked how much capacity the owner had - the answer was 24kWh. The Nissan rep was then asked the total pack capacity - the answer: That's Proprietary.

 

[vkruger]

I have over 300 cycles on my LiMn battery pack I built for my bike. When I researched the cells I found out that discharging to 3.0 V does no harm but charging above 4.1 V does. I charge my pack to only 4.06V. I would be surprised if Nissan allowed charging above 4.1 V. Data back then showed charging to even 4.15 V cut cycle life (to 80% of original capacity) in half. Discharging to 3.0 V had no measureable impact on cycle life.

I charge to 80% weekdays (because that is all I need) and 100% on weekend and don't worry one bit about loss of cycle life. I would prefer to charge to 95% all the time until I find out that 100% is less than 4.1 V, then I'll charge to 100% all the time. I think Nissan is just being over cautious.

 

[EVDRIVER]

I have over 300 cycles on my LiMn battery pack I built for my bike. When I researched the cells I found out that discharging to 3.0 V does no harm but charging above 4.1 V does. I charge my pack to only 4.06V. I would be surprised if Nissan allowed charging above 4.1 V. Data back then showed charging to even 4.15 V cut cycle life (to 80% of original capacity) in half. Discharging to 3.0 V had no measureable impact on cycle life.

I charge to 80% weekdays (because that is all I need) and 100% on weekend and don't worry one bit about loss of cycle life. I would prefer to charge to 95% all the time until I find out that 100% is less than 4.1 V, then I'll charge to 100% all the time. I think Nissan is just being over cautious.

Perhaps, but the have an 80% battery saver mode as well.

 

[IBELEAF]

Ugh, is there no way to charge to 95%

 

[charlie1300]

As a non-EE, I'm impressed with some of the technical arguments here that the Leaf's BMS will protect us from doing the battery pack damage. Yet I keep coming back to the fact that Nissan clearly recommends charging to 80% and not discharging below 20%. Given the fact that following their recommendation changes the Leaf from 100 miles/charge to 60 miles/charge, with a really negative marketing implication, I have to believe Nissan has some factual reason for the recommendation. I can't just write this off as corporate cautiousness.

 

[TomT]

As an EE, I have to adopt the position that if charging to 100 percent or discharging to "zero" was really such a significant consideration, they would never have allowed it in the first place.

As a non-EE, I'm impressed with some of the technical arguments here that the Leaf's BMS will protect us from doing the battery pack damage. Yet I keep coming back to the fact that Nissan clearly recommends charging to 80% and not discharging below 20%. Given the fact that following their recommendation changes the Leaf from 100 miles/charge to 60 miles/charge, with a really negative marketing implication, I have to believe Nissan has some factual reason for the recommendation. I can't just write this off as corporate cautiousness.

 

[evnow]

As a non-EE, I'm impressed with some of the technical arguments here that the Leaf's BMS will protect us from doing the battery pack damage. Yet I keep coming back to the fact that Nissan clearly recommends charging to 80% and not discharging below 20%. Given the fact that following their recommendation changes the Leaf from 100 miles/charge to 60 miles/charge, with a really negative marketing implication, I have to believe Nissan has some factual reason for the recommendation. I can't just write this off as corporate cautiousness.

Li batteries do best if kept in the middle of usable range (see Volt's usage of 10.4 kwh out of 16 kwh). It is also possible Nissan is using more of the topend than bottom end (or the otherway round).

Either way, ambient temperature and miles driven will have more of an impact than100% charging.

 

[AndyH]

I have over 300 cycles on my LiMn battery pack I built for my bike. When I researched the cells I found out that discharging to 3.0 V does no harm but charging above 4.1 V does. I charge my pack to only 4.06V. I would be surprised if Nissan allowed charging above 4.1 V. Data back then showed charging to even 4.15 V cut cycle life (to 80% of original capacity) in half. Discharging to 3.0 V had no measureable impact on cycle life.

I charge to 80% weekdays (because that is all I need) and 100% on weekend and don't worry one bit about loss of cycle life. I would prefer to charge to 95% all the time until I find out that 100% is less than 4.1 V, then I'll charge to 100% all the time. I think Nissan is just being over cautious.

Prepare to be surprised. Nissan's using a lithium-manganese variant and the numbers I reported in my diagram modifications (95% SOC 4.0-4.2V) comes directly from Nissan's service manual.

It's very important that we keep in mind that different cells have different voltage limits.

 

[AndyH]

As a non-EE, I'm impressed with some of the technical arguments here that the Leaf's BMS will protect us from doing the battery pack damage. Yet I keep coming back to the fact that Nissan clearly recommends charging to 80% and not discharging below 20%. Given the fact that following their recommendation changes the Leaf from 100 miles/charge to 60 miles/charge, with a really negative marketing implication, I have to believe Nissan has some factual reason for the recommendation. I can't just write this off as corporate cautiousness.

Can you reference the 20% recommendation? I don't recall reading that. Thanks!

 

[AndyH]

Li batteries do best if kept in the middle of usable range (see Volt's usage of 10.4 kwh out of 16 kwh). It is also possible Nissan is using more of the topend than bottom end (or the otherway round).

Either way, ambient temperature and miles driven will have more of an impact than100% charging.

I think that Nissan is using more of the 'top' of the charge. This is common practice for other lithium variants that carry more of their energy higher in the discharge curve, like LiCo. This 'top bias' seems to be confirmed by a number of government lab papers, including your recent find documenting the affect of HVAC on range.

(This paper also shows that the lithium nickel manganese variant has a bit higher voltage - both nominal and 100% SOC - up from the LiPo/LiMn2O4 4.2V to at least 4.3V - and that better fits the nominal voltage reported by AESC for Leaf cells, and better fits the voltages monitored by the Leaf's management system.)

I think this also allows us to use all the numbers we've been quoted - 80%, 95%, 24kWh, etc. It's clear that we have 24kWh of usable energy on the 'consumer side'. Top bias would fit well with 'consumer 100%' being 95% ultimate SOC. We'd still be using 80% of capacity by leaving 15% on the bottom.

This top bias also gives us (finally) a reason why charging to 80% 'consumer' would be a bit easier on the pack.

One thing's for sure - the 'zone of understanding' is moving around a bit as it expands.

 

[tbleakne]

As an EE, I have to adopt the position that if charging to 100 percent or discharging to "zero" was really such a significant consideration, they would never have allowed it in the first place.

As a non-EE, I'm impressed with some of the technical arguments here that the Leaf's BMS will protect us from doing the battery pack damage. Yet I keep coming back to the fact that Nissan clearly recommends charging to 80% and not discharging below 20%. Given the fact that following their recommendation changes the Leaf from 100 miles/charge to 60 miles/charge, with a really negative marketing implication, I have to believe Nissan has some factual reason for the recommendation. I can't just write this off as corporate cautiousness.

We should assume Nissan used their best engineering and economic judgement to pick the high and low hard boundaries beyond which the car will not push the battery pack. This judgement also figured into the warnings Nissan has placed in both the manual and the car itself to encourage us to use the extreme boundaries of the battery sparingly to avoid premature loss of battery capacity.

In picking these points cost and weight had to be traded off against range and battery life. There is no magical single "safe" point at either the high or low end. Instead the rate of decline of battery capacity is a complex function of its total history of charge and discharge cycles and temperature exposure.

Strain on the battery increases non-linearly as you approach both the fully charged and fully discharged points. Arriving home most of the time with at least two or more bars left of SOC is certainly better than arriving home frequently with only one bar left. Similarly charging to 90% SOC frequently when you don't need 100% will provide more than half the benefit of stopping charge at 80% SOC. Too bad Nissan did not provide a convenient 90% threshold. Maybe they can do that with a firmware upgrade.

 

[evnow]

I think this also allows us to use all the numbers we've been quoted - 80%, 95%, 24kWh, etc. It's clear that we have 24kWh of usable energy on the 'consumer side'. Top bias would fit well with 'consumer 100%' being 95% ultimate SOC. We'd still be using 80% of capacity by leaving 15% on the bottom.

It might even be 95% at the top and 10% at the bottom.

So, we get Battery Low at 25% and Turtle at 17%. The total battery size would be some 27.5 kwh.

 

[AndyH]

I think this also allows us to use all the numbers we've been quoted - 80%, 95%, 24kWh, etc. It's clear that we have 24kWh of usable energy on the 'consumer side'. Top bias would fit well with 'consumer 100%' being 95% ultimate SOC. We'd still be using 80% of capacity by leaving 15% on the bottom.

It might even be 95% at the top and 10% at the bottom.

So, we get Battery Low at 25% and Turtle at 17%. The total battery size would be some 27.5 kwh.

Could be, but the cells with cobalt tend to have their power near the top, so there's not much reason to work near the bottom. Plus, dropping to the last 10% would 'violate' that we have 80% to work with. (yeah, I know - like that's an absolute maybe at this pont. )

 

[tbleakne]

In other words, the bars have no strict relationship to SOC.

The 10th bar (third one down) will come on when the battery SOC crosses from 79% to 80% - but will turn off when the SOC crosses from 85% to 84%. So the meaning of the bars is relative to the direction.

I thought this was a typical way to engineer indicator lights to avoid flickering back and forth when close to between the levels.

The reason the SOC display sometimes changes when you switch to/from charging/driving is the battery's intrinsic hysteresis. Below is a graph which displays battery voltage vs SOC for both charge and discharge cycles for Li-ion chemistry.

Note how a given voltage has a substantially different SOC for charge and discharge. The Leaf system must compute the SOC taking this effect into account, but it can't correct perfectly, and that is why you see some changes in the display.

Credit: http://www.mpoweruk.com/chargers.htm
This site has quite a bit of valuable battery and charging info under the "Energy Storage" tab.