Nissan Leaf Cells Test Data
- Thread starter icecube57
- Start date
Hillhater
100 TW
Neat kart
..but please have the kids wear a helmet & gloves etc.
Thinks can go wrong very quickly on a kart !
..but please have the kids wear a helmet & gloves etc.
Thinks can go wrong very quickly on a kart !
inedible
100 W
Got around to taking one of these apart today, took some nice high res pictures of the process.
My goal was to leave the corners intact so the stack of modules can still be bolted together. So, unlike others, I left most of the can intact, just opening it enough to get at the tabs.
Here's a link to the album: http://imgur.com/a/ULQfj
I'm happy to say I didn't explode at all. Not even a bit.
My goal was to leave the corners intact so the stack of modules can still be bolted together. So, unlike others, I left most of the can intact, just opening it enough to get at the tabs.
Here's a link to the album: http://imgur.com/a/ULQfj
I'm happy to say I didn't explode at all. Not even a bit.
teslanv
1 MW
inedible said:Got around to taking one of these apart today, took some nice high res pictures of the process.
My goal was to leave the corners intact so the stack of modules can still be bolted together. So, unlike others, I left most of the can intact, just opening it enough to get at the tabs.
Here's a link to the album: http://imgur.com/a/ULQfj
I'm happy to say I didn't explode at all. Not even a bit.
NICE! That looks much cleaner.
inedible
100 W
Can anyone tell me what gauge of wire I should use to connect the cells?
I see in his videos that user mvly used two strands of 14awg between each cell, and said in theory 1 strand of 14awg is enough, but I'm not sure that's right. Why do they have such thick solid copper bus bars between the cells in the original configuration? Surely that's there for a reason.
I've only got 20awg laying around. I was thinking of using three or four strands or something, but what do you think? Am I better off buying some 12awg between the cells, maybe some 8awg for the main output?
I don't even really know what amperage my motor is pulling. It's rated at 500w, 48v, so that ought to mean it's using ~10A, yet my controller says 40A on it. The existing wires going off my stock 48v SLA battery are like 12awg I think.
I see in his videos that user mvly used two strands of 14awg between each cell, and said in theory 1 strand of 14awg is enough, but I'm not sure that's right. Why do they have such thick solid copper bus bars between the cells in the original configuration? Surely that's there for a reason.
I've only got 20awg laying around. I was thinking of using three or four strands or something, but what do you think? Am I better off buying some 12awg between the cells, maybe some 8awg for the main output?
I don't even really know what amperage my motor is pulling. It's rated at 500w, 48v, so that ought to mean it's using ~10A, yet my controller says 40A on it. The existing wires going off my stock 48v SLA battery are like 12awg I think.
teslanv
1 MW
inedible said:
Wire gauge size will depend on how many amps you intend to run through the controller. Unless you have a high-powered controller (>40A), 12Ga wires should be sufficient.
inedible said:
The reason why I went with dual 14 gauge is because it was what I had available. But like other said it depends on what you draw and how long your wires is from the battery to the controller.
I would say if you are keeping your power tamed, 12 gauge are typically OK.
dnmun
1 PW
i use 12AWG solid wire on my lipo pack and it has had 95A surges with no problem.
inedible
100 W
I made a diagram for the polarity of the cells in case anyone wants a reference while soldering. The green connections of course show how to make a 4S1P pack.
This also means if I had been paying attention before cutting, I could have skipped cutting this connection:
Oh well.
This also means if I had been paying attention before cutting, I could have skipped cutting this connection:
Oh well.
inedible
100 W
Pretty straightforward. I don't have a pic right now but I just dremeled away the plastic above it so I could get at the copper, and did the same cuts as the top.
I also dremeled away the plastic above where the negative post from the original terminals was. It's closer and easier to solder from that than the actual tab.
I also dremeled away the plastic above where the negative post from the original terminals was. It's closer and easier to solder from that than the actual tab.
Well 4300 miles and over 130 cycles so far at 1-2C discharge and much slower C charging and still going strong. It sags more now, but still gives the full 25Ah I was getting when I first got them used.
I don't think these are great for large C discharge. Too much sag and probably not good for the battery. If these were indeed put in power tools, I can see why they would not last long. That would be Milwaukee's failed engineering, not the battery chemistry itself.
I don't think these are great for large C discharge. Too much sag and probably not good for the battery. If these were indeed put in power tools, I can see why they would not last long. That would be Milwaukee's failed engineering, not the battery chemistry itself.
mvly said:
Can you quantify the sag? What voltage are you dipping to and what amperage for that voltage? I'm trying to get an idea of how bad these really are. Especially considering large format LiFePO4 like CALB sag to 2.889v with a 30 second 576 amp 8C draw on their 72Ah cells with a resting voltage of about 3.3v. I'm wondering if these Leaf are really only the budget conscious option to take.
MN Driver said:
To give you a rough idea, at 1C, I now get around 3-4V drop from a 24s pack. 1C for me is around 25-30A from my 25Ah pack. Remember I have the packs in 4s1p vs the stock 2s2p configuration and you can realistically get only 25Ah out of these. Lets say the pack was idling at 96V. When I pull 25-30A, the voltage goes down to around 92V.
When the pack was newer, when I got it, the voltage drop at 1C is around 2-3V.
The drop is pretty consistent throughout the voltage range. i.e. 3-4V whether the pack is full or empty.
In summary, when it's newer the sag would be roughly 0.125V at 1C per cell. Now it's 0.167V at 1C per cell.
At 2C, currently the sag is closer to 6-7V for the 24s pack whether full or empty. When it was newer, it would be around 5-6V
So close to 0.25V @ 2C per cell when newer and now it's closer to 0.3V @ 2C per cell.
So all in all, the sag is pretty tamed if you keep the current draw to 1-2C. I limit my max current draw to around 60A battery.
I expect these pack to sag more when winter comes around. By then I will have another 100-150 additional cycles on it bringing total usage to around 300 cycles. I'll update you guys once it gets there on the sag and performance. But I expect these cells to last a very long time. My guess, at the current rate of degradation, I expect it to get at least 600 cycles before I really have to retire the pack. But by then the sag will be pretty horrible and maybe the capacity will be much lower.
I am surprised the CALB LiFePo4 can source that much with minimal sag. But then again it's LiFePo4, which are known to hold voltage very well if used within it's power capacity. I am not sure if the LEAF cells are good for more than 2C current draw. I have not tested it yet, but I suspect it will sag quite a bit. If you really want 500-600A out of these, I recommend get more in parallel. Probably put 4 stock 2s2p packs in parallel to get 500A current draw with minimal voltage sag. You get more energy though in this configuration. 4 stock 2s2p packs in parallel will be 240Ah vs the 72Ah of your LiFePo4.
I have a LEAF, and the motor pulls 80kW+ at full throttle, so that is over 3C. Voltage does appear to sag pretty significantly, I'll try to get some data.mvly said:
Nissan does not limit power until well after the cells have lost 30% capacity - don't know if anyone's gotten them to lose much more capacity than that. Keep the cells as cool as you can, they don't like the heat. They'll sag more in the cold, but work fine down to freezing temperatures. Nissan heats the pack if they get down to -4F and then turns off the heater when the pack reaches 14F. Discharging is fine at high rates when cold, but be very careful charging when temps are below freezing.
What voltage limits are you using? Nissan shuts the car down when the resting voltage gets just above 3.0V resting voltage, 80% "long life" is around 4.02V / cell and "100%" is just over 4.1V / cell (all resting voltages).
drees said:
According the specs of the battery, full charge is 4.2V. However I am not aware of the 4.1V 100% capacity. The voltage does drop relatively fast from 100.3V to around 98V-99V just after 0.25Ah usage. It's good to know I should charge it to lower limits.
I use mine from 4.17V full charge down to 3.7V at the lowest. I took it down to 3.0V average at idle once, but the cells were all imbalance. I would keep these cells above 3.6V idle.
I think I did a quick calculation on this discharge stuff and concluded these cells can handle up to 5C burst if their specs were correct. Though I would think the sag at those current rate is pretty bad. I still think these are only 1-2C cells. I mean 1-2C is already 25-60A even if you convert these cells from 2s2p to 4s1p. Higher if you don't convert. And most people will probably not use it more than 60A continuous for the most part. If they really wanted power, I would recommend Lipo Nanotech with hefty wires to keep the weight down while still keeping it high power.
I think the only limiting factor for these pack right now is its size format. It's 25Ah minimum per cell so for those who only need 10-20Ah, you are out of luck. This is not for everyone. But building a 44V25Ah pack (12s1p) for general purpose pack should be ideal for everyone who wants to get into ebike with good range, low-mid power, and safe lithium chemistry. Definitely beats build battery packs from 18650 cells!
Yes, I think it's a good idea to limit your voltage a bit more. It will reduce your usable capacity now, but it should improve it down the road. How much imbalance are you seeing at lower SOC? Nissan also top-balances the pack. When charged to 100% balance is typically around 30mV or better ('13+ LEAFs seem to keep balance better), but as you note, once resting voltage gets down to 3.6V you can see 100mV difference. The pack is nearly fully discharged then, anyway.mvly said:
Common procedure for determining peak discharge rates is to do pulse discharges until the lowest cell reaches a minimum voltage, typically 2.5-3.0V or something. I think NREL testing showed that the LEAF pack will do 200 kW pulse discharge across most of the SOC range, that's around 8C. Of course, it won't do that very long. You should be fine at 3-4 C for pulse discharge rates, just don't let the cells drop below 3.0V. But then again even Nissan lets the cells drop below 3.0V under load at low SOC.mvly said:
drees said:
when it was 3V average per cell, I was getting major imbalance. It some were 2.8V. Other were 3.4V. Most were around 3V though. Some were at low as 2.7V. At 3.6V average, I was getting around 100mV imbalance. i.e. some were 3.5V other were 3.65V. At 3.7V, I didn't notice any imbalance. For instance, most were like 3.73V and maybe 1 or 2 were at 3.72V. So yeah they are pretty good in terms of staying balance.
I don't know if I want to charge to 4.1V. That would be 98.4V at full charge. Any going from 4.17V to 4.1V is at least 1Ah lost. It's not much, but still that is about 2-3 miles depending on how fast I ride. Maybe I will lower the end voltage to like 4.15V or 4.13V. Thanks for the recommendations.
Are you sure I should only charge to 4.1V? I checked the battery specs and it states 403.2V max. This is 96 cells in series charged to 4.2V each. Nominal is 360V which correspond to 3.75V per cell.
Heh, I don't keep it at 4.17V for long anyways. I usually charge to full and within 1-2 hours I ride. So i guess it doesn't really matter. The rest of the time it's mostly around 3.95V
I can't tell you how much of a difference in cycle life it will make going well above 4.1V, I can only tell you what Nissan does with the cells.mvly said:
FWIW, I grabbed a full-load (80kW indicated) data point for you from my LEAF this morning. Keep in mind the LEAF has 48 of these modules. Data is as transmitted over the car's CAN bus.
SOC: 66%
AHr: 53.4Ah
Resting: 382V (3.98V/cell)
80kW Load: 355V (3.70V/cell)
Pack Temp: 23-24C
That is around 225A being pulled from the pack and since it's a 2S2P setup that's about 112-113A per cell.
Hillhater
100 TW
do you have a reference to those CalB tests ?MN Driver said:
That is implying an IR of less than 0.001 ohm, which is much less than even the Nanotec lipo cells !
http://endless-sphere.com/forums/viewtopic.php?f=14&t=58210&p=906009#p906009
Or am i missing something here ?
zEEz
1 kW
Hillhater said:That is implying an IR of less than 0.001 ohm, which is much less than even the Nanotec lipo cells !
http://endless-sphere.com/forums/viewtopic.php?f=14&t=58210&p=906009#p906009
Or am i missing something here ?
the given sag value is valid between 29 and 30 seconds, because exactly at that time the boiling electrolyte cooked in the previous 29 seconds has reached the plasma temperature and the IR has become infinitesimal.
Poor cells :lol: they were intended for airplane energy backup use, no plasma ignition
beside that, 1 mohm of internal resistance could be realistic for a good cell, partially heated.
have fun
Skip to 1:17:30. Pay attention to the time he actually turns the load on and count 30 seconds because he actually runs out about 40 seconds or so.Hillhater said:do you have a reference to those CalB tests ?MN Driver said:
That is implying an IR of less than 0.001 ohm, which is much less than even the Nanotec lipo cells !
http://endless-sphere.com/forums/viewtopic.php?f=14&t=58210&p=906009#p906009
Or am i missing something here ?
https://www.youtube.com/watch?v=UCPjW57sr7w
