Cycle life tests of High Energy density cylindrical cells
- Thread starter Pajda
- Start date
Pajda
10 kW
john61ct said:
These results are in updated graphs and table in the first post. Particularly are added charts after 1000 cycles of voltage drops and temperature rise. it is also interesting to point out that the DCIR values and the associated temperature rise after 1000 cycles are slightly better than on the brand new cell. This is a fairly common phenomenon in original chinese cell development.
Pajda
10 kW
small update after a while
added more graphs representing also 3C discharge rate for NEW and after 1000 CYCLE (both voltage drops + temp. rise)
added more graphs representing also 3C discharge rate for NEW and after 1000 CYCLE (both voltage drops + temp. rise)
Pajda
10 kW
small update after a while
LG M50LT recently finished 1000 cycles. Due to the vacation time it will take another week or two to run fInal test on precision tester to get data for update of charts and table.
LG M50LT recently finished 1000 cycles. Due to the vacation time it will take another week or two to run fInal test on precision tester to get data for update of charts and table.
Pajda
10 kW
small update after a while
updated values for LG M50LT after 1000 cycles and Samsung 50E2 graphs
and small teaser...
This week I found LG INR21700 M58T (5570 mAh) cell on Nkon store. It is quite new cell which can finally beat the magical border of 800 Wh/l in cylindrical format. Samples are on the way to me and so I hope that the standard cycle life tests will start later next week.
updated values for LG M50LT after 1000 cycles and Samsung 50E2 graphs
and small teaser...
This week I found LG INR21700 M58T (5570 mAh) cell on Nkon store. It is quite new cell which can finally beat the magical border of 800 Wh/l in cylindrical format. Samples are on the way to me and so I hope that the standard cycle life tests will start later next week.
Pajda
10 kW
small update after a while
LG INR21700 M58T (5570 mAh) cell form Nkon store added to the chart and table. This cell sample was removed from modules, but the welds are nicely cleaned. This sample was tested as "naked" without tube.
Unfortunately, due to the larger dimensions D: 21.7 mm and H: 71 mm, the volumetric energy density is only 768 Wh/l and therefore the "magic" value of 800 Wh/l in cylindrical format still resists.
LG INR21700 M58T (5570 mAh) cell form Nkon store added to the chart and table. This cell sample was removed from modules, but the welds are nicely cleaned. This sample was tested as "naked" without tube.
Unfortunately, due to the larger dimensions D: 21.7 mm and H: 71 mm, the volumetric energy density is only 768 Wh/l and therefore the "magic" value of 800 Wh/l in cylindrical format still resists.
goatman
10 MW
the m58 seems to be burning off capacity into heat when comparing to the m50lt at higher discharge rates
Pajda
10 kW
small update after a while
updates values and charts for BAK N18650-35CP after 1000 cycles (Data for new cell will be added soon). Also Panasonic NCR18650T1 finished 1000 cycles and data will be updated soon as well.
LG INR21700 M58T (5570 mAh) updated data after first 50 cycles. I have added the data of two more cycles after 50th nominal to show that we will be observing a decent "sawtooth" trend. I'm not surprised by the behavior that cell does not like cycling at lower charging voltages, I've just never seen it this strong within LG production. Next to this 95% DoD test (4.15-3.00V) the standard 100% DoD test (4.20-2.50V) runs in parallel and here the cycle life looks very good so far.
updates values and charts for BAK N18650-35CP after 1000 cycles (Data for new cell will be added soon). Also Panasonic NCR18650T1 finished 1000 cycles and data will be updated soon as well.
LG INR21700 M58T (5570 mAh) updated data after first 50 cycles. I have added the data of two more cycles after 50th nominal to show that we will be observing a decent "sawtooth" trend. I'm not surprised by the behavior that cell does not like cycling at lower charging voltages, I've just never seen it this strong within LG production. Next to this 95% DoD test (4.15-3.00V) the standard 100% DoD test (4.20-2.50V) runs in parallel and here the cycle life looks very good so far.
Pajda
10 kW
small update after a while
BAK N18650-35CP and Panasonic NCR18650T1 finalized data in the charts
Samsung INR21700-50G added to the test.
Comment on the measured temperatures: here I have to apologize that I missed adding important information to the results
*Temperature is measured by the K-type thermocouple, attached to the cell body using Kapton tape. Some cells are sold as Naked (without tube) and these can show significantly higher temperatures than cells with tube due to the effect of the tube thermal insulation.
How much difference there is can be seen between cells 50E2(with tube) and 50G(naked), which otherwise show almost identical parameters. So even the closely watched M58T is now at a disadvantage against the others because it was measured naked as well.
BAK N18650-35CP and Panasonic NCR18650T1 finalized data in the charts
Samsung INR21700-50G added to the test.
Comment on the measured temperatures: here I have to apologize that I missed adding important information to the results
*Temperature is measured by the K-type thermocouple, attached to the cell body using Kapton tape. Some cells are sold as Naked (without tube) and these can show significantly higher temperatures than cells with tube due to the effect of the tube thermal insulation.
How much difference there is can be seen between cells 50E2(with tube) and 50G(naked), which otherwise show almost identical parameters. So even the closely watched M58T is now at a disadvantage against the others because it was measured naked as well.
Pajda
10 kW
Samsung 48x is ready for the test, but it will be overtaken by the second HE competitor with more than 5300 mAh capacity in 21700 format.
Pajda
10 kW
small update:
Still waiting for the BAK CD with 5400 mAh in 21700 format.
Also small comment to the LG M58T results which recently pass 500 cycles under 100% and 95% DoD. This cell with nominal energy density of 800 Wh/l (naked cell, ca 770 Wh/l with tube) turns a design of BEV with 400 miles or 640 km of WLTP range into a child's puzzle. The 500 cycles at 100% DoD (SoH: 88% of nominal capacity) gives you 320.000 km of vehicle mileage. It will be beneficial to check how the cell stands from a fast charging perspective.
Still waiting for the BAK CD with 5400 mAh in 21700 format.
Also small comment to the LG M58T results which recently pass 500 cycles under 100% and 95% DoD. This cell with nominal energy density of 800 Wh/l (naked cell, ca 770 Wh/l with tube) turns a design of BEV with 400 miles or 640 km of WLTP range into a child's puzzle. The 500 cycles at 100% DoD (SoH: 88% of nominal capacity) gives you 320.000 km of vehicle mileage. It will be beneficial to check how the cell stands from a fast charging perspective.
Pajda
10 kW
another update:
Thanks to Tumich (ES member) I got samples of LGX E78 (78 Ah) pouch cells which are used in several EV platforms. One of them is VW MEB platform, so you can find them in VW ID.x or Škoda Enyaq family vehices (but recently they start to use also prismatic cells from CATL, so it is hard to tell).
You can find data in third post of this thread how it stands against the cylindrical cells. Unfortunatelly that is all for near future as I do not have free high power channel for cycle life test.
Thanks to Tumich (ES member) I got samples of LGX E78 (78 Ah) pouch cells which are used in several EV platforms. One of them is VW MEB platform, so you can find them in VW ID.x or Škoda Enyaq family vehices (but recently they start to use also prismatic cells from CATL, so it is hard to tell).
You can find data in third post of this thread how it stands against the cylindrical cells. Unfortunatelly that is all for near future as I do not have free high power channel for cycle life test.
Pajda
10 kW
john61ct said:
The right question is who doesn't use modules from CATL today... Right now VW use CATL NMC prismatic cells with the comparable energy density as LG pouches. VW MEB modules was designed with intention to use both formats. Prismatic cells for MEB modules are marked as L221 size (the older VDA-PHEV2 format is then marked as L149) where the L221 size cell dimensions are ca 221 x 33.6 x 106 mm. There are already several manufacturers of cells in this particular format (CALB, EVE..) In actual VW MEB modules from CATL is used 16 cells in 8s2p connection with ca 117Ah capacity per cell. This is equal to LG modules 8s3p with E78 pouches with 78Ah capacity. Both are used for the actually biggest 82kWh MEB platform battery pack.
As for LFP, i think that VW is still not using this chemistry but is working on that.
j bjork
1 MW
Very interesting with the LGX E78 cell test, what do you think of the result?
It seems to sag a lot, I wonder how it would do on higher loads :?
It seems to me like vw uses it with a max of maybe 3c load, but already that seems to be much maybe?
It seems to sag a lot, I wonder how it would do on higher loads :?
It seems to me like vw uses it with a max of maybe 3c load, but already that seems to be much maybe?
Pajda
10 kW
My guess is that the dominant lenght of this particular format (i.e. internal current path between terminals) will have effect on voltage sag in comparison with small cylinders or short pouches.
But I think it is not a big problem in this use case. These E78 cells are mostly used in 82 kWh battery packs, where the MEB drivetrain power is 150 kW peak (220 kW in RS version). So 2C(3C) peak discharge for few seconds. The average discharge rate is ca 0.2C, where the E78 results are not bad.
The motivation for using large batteries in BEVs is that you spare the battery life during discharge (by lowering average loads and especially the DoD) and what you save here then can be sacrificed for fast battery charging.
But I think it is not a big problem in this use case. These E78 cells are mostly used in 82 kWh battery packs, where the MEB drivetrain power is 150 kW peak (220 kW in RS version). So 2C(3C) peak discharge for few seconds. The average discharge rate is ca 0.2C, where the E78 results are not bad.
The motivation for using large batteries in BEVs is that you spare the battery life during discharge (by lowering average loads and especially the DoD) and what you save here then can be sacrificed for fast battery charging.
Attachments
Pajda
10 kW
Temporarily added cell DMEGC INR18650-26E in the 95-100% DoD chart.
In the future(Christmas is coming) it will be moved to my third cycle life thread here on ES which will be called something like "Best value cells". The price per kWh with decent cycle will be the main parameter. Good global availability from a reputable supplier will be also an important parameter.
In the future(Christmas is coming) it will be moved to my third cycle life thread here on ES which will be called something like "Best value cells". The price per kWh with decent cycle will be the main parameter. Good global availability from a reputable supplier will be also an important parameter.
Hi, great thread. Which cells would you recommend for my 12S6P pack?
I am looking for maximum 1000 or at least 500+ cycles capacity with good price–performance ratio.
Max 0.5C, mostly 0.2C discharge current for my application.
Samsung INR21700-50E 4900mAh 9.8A Reclaimed €3.10
Samsung INR21700-50G 4850mAh 9.7A NO WRAP €3.40
LG INR21700-M50LT 4890mAh 14.4A €4.06
Are there any new better cells coming to the market soon?
I am looking for maximum 1000 or at least 500+ cycles capacity with good price–performance ratio.
Max 0.5C, mostly 0.2C discharge current for my application.
Samsung INR21700-50E 4900mAh 9.8A Reclaimed €3.10
Samsung INR21700-50G 4850mAh 9.7A NO WRAP €3.40
LG INR21700-M50LT 4890mAh 14.4A €4.06
Are there any new better cells coming to the market soon?
Pajda
10 kW
silenec said:
As you can see from charts, there is no significant difference in cycle life of this three competitors (and others from Lishen, BAK..). Maybe in Samsung production check the actual model available like 50E vs 50E2. Original 50E seems to have slightly better cycle life but higher DCIR than 50E2.
If you look for something significantly better, there is definitely one champion LG M58T with 5600mAh capacity and very good cycle life and DCIR. Unfortunately my guess is that M58T will not be generally available for about a year and the price will also stay unreasonable for some time. Another new model BAK 21700CD with 5300mAh capacity is now under testing.
Thanks! Then there is also this Samsung INR21700-48X 4650mAh rated at 3000 cycles @ 70% capacity. Did you test it by any chance? I wonder how it would compare with the other 1000 cycle @ 80% capacity rated cells at 1000, 2000 and 3000 cycles. After how many cycles do the 1000 cycles rated cells normally die / become unusable (unbalanced or <60% capacity)?Pajda said:
Cycle ratings are arbitrary, completely depend on the lab test parameters.
Can be easily tripled depending on how gentle the use case and a diligent owner's care factors.
Or with "abusive" use cases or ignorant/careless owners, even 50 cycles might be pushing it.
Can be easily tripled depending on how gentle the use case and a diligent owner's care factors.
Or with "abusive" use cases or ignorant/careless owners, even 50 cycles might be pushing it.
