Panasonic/Sanyo NCR20700B vs Samsung INR21700-50E

flippy said:
you are not going to get any decent cycle life on any of the 2xx cell form factors, not yet anway. if you want hard cycle life you need to dial it back to 18650 and stay under 3000mAh.

What about Samsung 30T 21700 : 3000 mAh rated 35A
and the Samsung 40T 21700: 4000 mAh rated 30A...
 
Matador said:
flippy said:
you are not going to get any decent cycle life on any of the 2xx cell form factors, not yet anway. if you want hard cycle life you need to dial it back to 18650 and stay under 3000mAh.

What about Samsung 30T 21700 : 3000 mAh rated 35A
and the Samsung 40T 21700: 4000 mAh rated 30A...

official ratings ignore lifespan, read the fineprint in datasheets, lifecycle is based on VERY low drains and charge cycles. just because the cell wont explode does not mean you wont kill it real fast.
if you use a 21xxx cell on those currents i expect 150 cycles or less before its basically dead.
 
Agreed
My VTC4 are rated for 30A, but I don't wan't to use them over 10A.

So for the Samsung 40T, same thing, would'nt push past 10A.

I don't think anybody ever got 300 cycle by pulling 30A continous out of VTC4 18650's in a large pack either. Plus the VTC4 were tested for cycle life with a 75 degree celcius cutoff, so it's not a REAL continous.

Let's be real here, nobody can ever expect datasheet stated cycle life when cell is continuously tested at maximum discharge rate (and regardless of cell temps).

Running the VTC4 at 30A continuous for 100% to 0%... For 300 cycle ? LOL !... Just look at the temp values at 30A and fully discharged.... Indeed, amp ratings are to be taken with a grain of salt....

vtc4 graph2.jpgSony US18650VTC4 (Rated 30A, 2100 mAh).jpg

But the Samsung 40T clearly is more suited than the Samsung 50E in higher discharge situations nonetheless... Such as low weight EV's

Matador
 
Still, the samsung 40T kicks ass.

60% residual capacity after 250 cycle when discharging it at 35A!!, without temperature cutoff....
Capture.JPG
I mean, they sure as hell last reall long if you only pull 10A/cell on them from time to time...

Wheras I really would'nt bet anything on the samsung 50E at 10A...

Matador
 
Yes inherent in such use patterns,

high-C **and** high-DoD discharge

often high-C charging too, right up to top voltage spec'd rather than sacrificing a few % SoC

some letting them sit there Full for days

coupled with a chemistry that gets pretty low lifecycles best of times

150-300 cycles is pretty darn good

Soon as anyone starts advising how to get longer life, people start objecting don't bother, who cares, convenience is more important etc.

Which is perfectly valid for those making their choices fully informed.
 
I think the INR21700-40T is very interesting in small dimensions of batteries like 10S2P, or 10S3P.
I daubt it, if the 50E, or M50 will have more Wh at this configuration on 250W-350W motors?!
 
Easy_Rider said:
I think the INR21700-40T is very interesting in small dimensions of batteries like 10S2P, or 10S3P.
I daubt it, if the 50E, or M50 will have more Wh at this configuration on 250W-350W motors?!

I agree...
Even if the 50E is supposed to be a 5000 mAh 21700 cell, that's probably only true at very low discharge current (like 0.2C or lower).

I mean look at this: https://zeroair.wordpress.com/2018/11/13/liionwholesale-samsung-50e-21700-li-ion-cell-review/
sans-titre.png

At 10 amps, you get massive sag, and only extract 4200-4400 of the claimed 5000 mAh, and that's on a brand new cell that wasn't yet cooked....
At 10 amps, the voltage sag is massive (due to internal resistance delta V= R x I), and the cell will heat and slowly cook... At 10A, the 50E capacity will rapidly degrade cycle after cycle, and you probably be getting much lower than 4200 to 4400 mAh by that time..

I would much rather invest in the Samsung 40T, they might only have 4000 mAh, but at 10A, you get at least 3900 mAh real, and they won't cooke as much,.

And also, factor in the BMS low voltage cutoff of (let's say 3.3V per cell) and Controller low voltage cutoff, and you really wont extract much real capacity with the samsung 50E, because they sag so much in voltage under load...
For example if you look at capacity at 10A up until you reach 3.3 Volts (LVC) on the Samsung 50E dicharge curve, and you only really extract 2000 mAh of capacity out of the claimed 5000 mAh. Not good...

I see the 50E being used in minimum 10S10P packs for ebikes that pull max 20A... but that's a 7 kilograms (15.4 pounds) battery at least to lug around...

Matador
 
Just few remarks :

1) Zeroair finished measurement at 2,8 V. Discharging down to 2,5 V according specification is giving capacity nearly 5 000 mAh in the range 1 – 5 A. ( equipment : calibrated ZKETECH EBC A-10H, BF-2A fixture, 25 ±1 °C)


Samsung 21700 - 50E f.jpg


2) Frankly, Zeroair measurement seems to be quite inaccurate.

3) Long term testing shows very good life time :

„Actually Samsung 50E seems to be very good choice. I did not ran the entire cycle life test, but I already did 500 cycles in some tests. I am surprised with the test results so far, because it shows much better results than 35E in 18650. It seems it has the same performance in comparison with LG M50 "B-grade" under low loads (up to 1C discharge) and much better performance under high load (3C) and fast charging (1C) torture tests. (I do not recommend to use it in high continuous load applications). But I have to apologize to ES community, that I still didn't start testing of the LG M50 "A-grade" cells. I suppose that "A-grade" M50 at least equals the 50E. And also M50 is after the recent price increase still about 10% cheaper than 50E.“

(source : Pajda)

https://endless-sphere.com/forums/viewtopic.php?f=14&t=98885&p=1450798#p1450798


Conclusion : Samsung 50E seems to be very good option, supposing current 5 A max.


Docware
 
I agree with you as well Matador! :wink:
The reason, why I bought a frog battery, based on the 21700 M50 from LG Chem, is because I use at 48V a motor (Tongsheng TSDZ2) with 250W. On 13S3P I have around 1,80A - 2,20A CDR on every cell...and I ride my bike in the city...and it is really flat (Hannover).
 
Are you sure that you agree with Matador ? 😊
Just read carefully his post. 😊


Docware
 
docware said:
Just few remarks :

1) Zeroair finished measurement at 2,8 V. Discharging down to 2,5 V according specification is giving capacity nearly 5 000 mAh in the range 1 – 5 A. ( equipment : calibrated ZKETECH EBC A-10H, BF-2A fixture, 25 ±1 °C)


Samsung 21700 - 50E f.jpg


2) Frankly, Zeroair measurement seems to be quite inaccurate.

3) Long term testing shows very good life time :

„Actually Samsung 50E seems to be very good choice. I did not ran the entire cycle life test, but I already did 500 cycles in some tests. I am surprised with the test results so far, because it shows much better results than 35E in 18650. It seems it has the same performance in comparison with LG M50 "B-grade" under low loads (up to 1C discharge) and much better performance under high load (3C) and fast charging (1C) torture tests. (I do not recommend to use it in high continuous load applications). But I have to apologize to ES community, that I still didn't start testing of the LG M50 "A-grade" cells. I suppose that "A-grade" M50 at least equals the 50E. And also M50 is after the recent price increase still about 10% cheaper than 50E.“

(source : Pajda)

https://endless-sphere.com/forums/viewtopic.php?f=14&t=98885&p=1450798#p1450798


Conclusion : Samsung 50E seems to be very good option, supposing current 5 A max.


Docware

I'd be curious to know... Is the low voltage cutoff of your BMS also 2.5 V. What about the low voltage cutoff of your eBike controller. Surely the 50E is a great cell, just not for eBike application... I don't know any eBike with a LVC that's even remotely close to 2.5V anyways.... More likely around 3.0 to 3.2V...
 
Easy_Rider said:
I agree with you as well Matador! :wink:
The reason, why I bought a frog battery, based on the 21700 M50 from LG Chem, is because I use at 48V a motor (Tongsheng TSDZ2) with 250W. On 13S3P I have around 1,80A - 2,20A CDR on every cell...and I ride my bike in the city...and it is really flat (Hannover).

1,80A - 2,20A CDR on each cell seem to be reasonable discharge rate for your application.
I'd be tempted to ask, what is the LVC (low voltage cutoff) set at on your ebike? In other words, what's your battery voltage when you battery is dead empty and your ebike won't run anymore ?

If it's dead at 41 V, then the LVC is 3.15V.... If it's dead at 43V, then the LVC is 3.31 V.
That would (according to docware's more accurate graph :mrgreen: ) give you either 4500 mAh (13.5Ah in 3S) or 4200 mAh (12.6 Ah in 3S) usable capacity respectively, assuming you pull an average 2A per cell...
 
docware said:
Are you sure that you agree with Matador ? 😊
Just read carefully his post. 😊
Docware

Of course I did and if you read my post, you can see my explanation.
I fit my battery to the Motor and controller I use and to the landscape and season I´d like to drive. And as long as I am 80 Kg. battery + motor are around 7 Kg. more, and I just ride for fun (no heavy weights in addition) on very easy landscape...the battery cells in my battery are not challenged in a way, that the voltage will drop down so much.
The controller has a cut-off at around 39V - 39,5V.
Few weeks ago I gave a friend back a cargo-Bike I converted with 48V system (also Tingsheng). In her case I used a 13S5P with 30Q, because her bike his heavy, she drives 2 small children every day and she uses the cargo-Bike as car replacement for shopping etc.
 
Easy_Rider wrote:

„I would not exceed, if I were you, the 0.3C for the battery.“ (charging)
„I fit my battery to the Motor and controller I use and to the landscape and season I´d like to drive“
„On 13S3P I have around 1,80A - 2,20A CDR on every cell.“


You obviously know a little bit about Li-ion cells 😊
I mean it´s nice to meet here somebody who is using his comprehensive knowledge intelligently.

Congratulation to your LG M50 choice.


Docware
 
I used to be a passionate RC-Freak and gained some experience on Li-polymer at the time, where 18650 could reach maybe 4C the most. :mrgreen: I remember trying to assemble a battery with CGR18650CH for some races...that was really desperating...
But in the meanwhile we have battery cells out there, that can really give you lot´s of power and also some reasonable energy density.
My first eBike was 8fun motor for the front wheel and the battery was Li-Polymer collection :mrgreen:
I am an Easy rider and an Easy learner :lol:
 
Yes, fortunately now we have cells like Samsung 50E or LG M50. LG maybe a little bit better, but generally both very similar and interesting for ebikes. :thumb:
 
Better, as in longevity, or maximum discharge rate?

I believe the two goals are opposed, so need to define "better".

Energy density is a third, but I guess an assumed constraint for both bikes and RC, which throws thermal safety out the window.
 
Better, as capacity at 2 A discharge for example. But I would rather say very similar.


Samsung 50E versus LG M50.jpg
 
Maybe the problem is "padja" is using a different name?

And I do not see any test showing a difference in cycle longevity between the various batteries being discussed in this thread.
 
Similar, not different :

https://endless-sphere.com/forums/viewtopic.php?f=14&t=98885&p=1450798#p1450798

By the way, Sanyo 20700B as younger, but bigger brother of Sanyo GA is also interesting cell for ebikes with its capacity 4100 mAh and 25 miliohm DCIR at 3,8 V/25 °C.
 
Back
Top