Panasonic NCR18650G - the mysterious 3600mAh battery's capacity test

Original article: https://www.thunderheartreviews.com/2019/02/panasonic-ncr18650g-mysterious-3600mah.html

Panasonic NCR18650G is a famous Li-ion cell which appeared in 2014 and very quickly disappeared. It had (and still has) the highest capacity rating in 18650 format.


The battery was bought from my reliable supplier (Queen Battery) and tested with ZKETECH EBC-A20 and a self-made battery holder. It's a PC-connected battery tester supporting 4-wire measuring and discharging at up to 20A.


I've used version 3.0 of my battery holder based on 0.5mm thick pure copper terminals



I've followed all the prescriptions of the IEC61960-2003 standard concerning battery's capacity measurement. Before each discharging cycle each battery was charged at standard charge current mentioned in its datasheet to charge end voltage. Before each discharging or charging i've held a 1-1.5hrs pause. The environment temperature was 23.0-24.5°C. To be sure in results i've done each test minimum twice (usually 3-4 times).


Panasonic NCR18650G

The cell is marked as E BV 8307. The 8307 code means that the cell was produced on March 7 2018.


I didn't find any datasheet of NCR18650G except this screenshot of a page from a Panasonic roadmap:


I compared it to other Panasonic cells' datasheets and made up these specs for NCR18650G:
Rated capacity: 3400mAh at 20°C
Minimum capacity: 3450mAh at 25°C (should be used as 1C)
Typical capacity: 3550mAh at 25°C
Nominal voltage: 3.6V
Standard charge current: 1035mA (0.3C)
Charge end voltage: 4.2V
Charge cut-off current: 69mA (0.02C)
Discharge cut-off voltage: 2.5V
Maximum weight: 47.5g

The measured DC IR of fully charged cell at 3.45A was 38±1.5mΩ
My cell's weight was 46.39g


Capacity test results:

At 0.2C / 0.69A NCR18650G's capacity was slightly higher than the 3550mAh declared and it's energy was almost 13Wh. At 5A discharge rate the capacity was a bit higher than 3350mAh and the energy - 11.2Wh. Most of the suppliers mention NCR18650G's max discharge current as 4.87A so i didn't go above 5A.

Compared to LG M36 and Samsung 36G (test), Panasonic NCR18650G looks better but i used 1.8A to charge the first two vs 1.04A in case of Panasonic.

Also the 0.3C standard charge current makes me think that NCR18650G would not have a great cycle life but it's just a suggestion.

Here is the video version of this review:


Check out my YouTube channel for batteries, chargers and other stuff reviews.
I've launched my blog where you can find all my reviews in one place. Every new test/review will be first published on YouTube and in the blog. I'll be happy to see new subscribers, comments, suggestions and just your thoughts.

I am starting to get mildly interested in these little cylindricals, coming from an LFP prismatics background.

What would it cost to get to say 100Ah @24-28V for just the cells?

New and from a highly trusted source to avoid QA rejects of course.

And very curious why these reviews never mention the specific chemistry used?

Li-ion is just a general umbrella term, as opposed to NMC or LMN or LCO. . .

Or are they all the same?

Are there recommended pack builders so I would not need to DIY?

Sorry if this is a derail, I can start a noob thread

Significantly different results to other tests on the same cell..
https://lygte-info.dk/review/batteries2012/Panasonic%20NCR18650G%203600mAh%20(Green)%20UK.html

John61....The Panasonic cells are NCA chemistry.

And I see not available, at least not new from trusted sources?

thunderheart said:

The measured DC IR of fully charged cell at 3.45A was 38±1.5mΩ
My cell's weight was 46.39g

Click to expand...

From your discharge graph, at around mid-capacity, going from 0.69A discharge to 5.00A, you can see the voltage drops from 3.60V to 3.26V with the added current.

That would be a DC IR more in the range of 79 mΩ... No where near that claimed 38±1.5mΩ DCIR (aka 36.5 to 39.5mΩ). I dunno where that 38 mΩ number came from, but it would be very surprising for such a low current rated cell to have such an okay range DCIR.

Demonstration of calculation, from looking at your above discharge curves (DCIR at 1980mAh State of Discharge, aka approx 55% SoD):
DC IR (@ 55% SoD) = (3.60V - 3.26V) / (5.00A - 0.69A) = 0.34V (drop) / 4.31A (add. current) = 0.079 Ω = 79 mΩ

If you redo the calculation further down after more discharge, the DCIR gets even worst (ex: 2880 mAh SoD; aka aprox 81% SoD)
DC IR (@ 81% SoD) = (3.38V - 2.98V) / (5.00A - 0.69A) = 0.40V / 4.31A = 0.093 Ω = 93 mΩ

Even at the start of the discharge cycle, it does not even get close to the claimed 38 mΩ (ex: at 180 mAh SoD; aka approx 5% SoD):
DC IR (@ 5% SoD) = (4.04V - 3.75V) / (5.00A - 0.69A) = 0.29V / 4.31A = 0.067 Ω = 67 mΩ

So, for sure these DCIR numbers are way far off the 38 mΩ claimed DCIR. So either the discharge curves observed are flawed by unnacounted additionnal IR from the measurment wirering harness, or the claimed 38 mΩ DCIR is a lie. Or the 38 mΩ is actually not the DCIR but rather the 1.0 kHz AC internal impedance (aka AC IR @1.0 kHz).

EDIT: Could be due to your wires... As good as your 101% IACS copper contacts are, the aligator clips are clearly the current bottleneck here, adding significant DCIR IMHO. Even if you solder the copper terminals directly to the wires (to get rid of the aligator clips), the Lead/Tin solder is a poor conductor... You would need proper crimping instead of soldering and massive wire gauge with no bottleneck, with the shortest possible lenght... And 4 wire measuring device with seperate measuring points for voltage drop measurement and independant seperate current draw points.

Matador

Something like this, but with copper and no soldered bottleneck:

Source: https://www.arbin.com/store/battery-holders/high-current-battery-holder/
Holder explained more here: https://www.arbin.com/news/page/2/

See the massive brass nuts and bolts on top... that's where you attach your current load leads with properly crimped copper ring terminals... So NO solder here.

Matador

Also look at how DCIR varies with State of Charge(SoC)/State of Discharge (SoD) and temperature.
You have to account for that. An ES member already showed that in the past, quite nicely too:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=73701

I must say all your efforts are really appreciated. I just hope you continue to refine your testing methods, because all this data is very usefull to us EV builders. The cell manufacturer data provided in spec sheets is far too incomplete as you already know.
Thanks


BTW, my vision of cell testing, in graphical terms:
Matador

Hillhater said:

Significantly different results to other tests on the same cell..
https://lygte-info.dk/review/batteries2012/Panasonic%20NCR18650G%203600mAh%20(Green)%20UK.html

Click to expand...

Yep.
The cell i've tested showed the results i published. And the difference gonna be even more if we take .dat file and calculate the results in Excel, because EB Tester Software calculates in his own way resulting in lower total values of capacity and energy.

Matador said:

I dunno where that 38 mΩ number came from

Click to expand...

From EB Tester Software's Resistance test. Here is the full series of DCIR values (minimum 1 min pause between tries) measured on a fully charged cell (1hr pause) at 3.45A: 37.1, 39.4, 39.1, 38.3, 36.5, 38.0, 37.1, 38.6, 37.7, 38.0, 39.7, 38.8, 37.1

thunderheart said:

From EB Tester Software's Resistance test. Here is the full series of DCIR values (minimum 1 min pause between tries) measured on a fully charged cell (1hr pause) at 3.45A: 37.1, 39.4, 39.1, 38.3, 36.5, 38.0, 37.1, 38.6, 37.7, 38.0, 39.7, 38.8, 37.1

Click to expand...

Seems like the this EB Tester produces DCIR results with high precision, but poor accuracy then.... because the same discharge curves produced by that same EB Tester tells otherwise in terms of DCIR values, as I demonstated from calculations I showed above.

Matador