Samsung INR21700-48G

[Hillhater]

Have you seen other tests on these cells ?
Here is one such, on this forum.. https://endless-sphere.com/forums/viewtopic.php?f=14&t=91926
It fairly clear from those tests that the full 4800mAh iis available , but there is very little between 4.2-4.1 volts (350mAh ?). And less in the 2.9 - 2.5 volt range (200mAh)
And here the test only ran to 2.8 v cut off, but still discharged over 4700mAh
http://lygte-info.dk/review/batteries2012/Samsung%20INR21700-48G%204800mAh%20(Gray)%20UK.html
So , run your 4.2-2.5 volt discharge test, then compare results and test methods.

 

[ElectricGod]

Life has been busy!

I have done discharge tests on several sets of cells now down to 2.5 volts from about 4.1+ volts and I'm consistently seeing 4400mah or less.

I have tried load testing at about 11 amps, 6 amps and a few lower current draw amounts. 4400mah is about all I ever see. I'm recording my results as I get them and will post that later. I'm not overly excited about these cells! Charging from around 2.5 volts to full typically gets me 4400mah or less.

 

[ElectricGod]

Well I am done testing INR21700-48G cells!

The INR21700-48G is disappointing. The best I have gotten is 4400mah and typical has been 4200mah. I've never gotten 4800mah and IR is pretty high. I seriously doubt I will buy more of this cell. I've now charged and discharged every one of the 40 cells I have on my SkyRC charger and a Chargery 12S balance charger twice each per cell and get similar results. I've also used a watt meter and my halogen bulb load...same thing. My test results are 100% less than acceptable. Samsung is over rating these cells. That or I got 40 sub par cells!

IMHO...buy something else...the 48G is at best a lack luster cell. MEH!!!

 

[Hillhater]

Did you compare your test method and equipment with others who have achieved satisfactory results ?
Have you checked /recalibrated your measuring equipment ?
Did you run control tests on other known capacity cells ?
Its not likely that other reputable tests have falsely achieved spec results, and equally unlikely you have received 40 bad cells. ?
So you have to suspect an error in your equipment or test proceedure.

 

[TheBeastie]

That is a pretty good cell, but after looking around online it seems pretty expensive for the pure Ah/$.
If you didn't look at it or know it was a 21700 cell you wouldn't be so interested.

I was thinking the cell is kind of similar to my current battery pack of Samsung 29E cells which are also quite a low C-rate but were excellent value at the time I bought them, when I first started testing a battery pack out with them I found 4P was not acceptable for me on a 500W Bafang ebike. So I bought more cells and made a 7P pack, this has worked out well as I only charge at best once a week. Being doing a lot of lazy summer rides this summer and have managed to get 100KM range out of the pack due to going slow.

Looks like this INR21700-48G is kind of the next generation of such a pack but it needs to be cheaper in terms of pure Ah/$.

 

[youyoung21147]

@ElectricGod : your conclusion reach mine ! These cells are overpriced and useless.

I think they are 35E chemistry fitted in a larger case : when you normalize numbers and curves proportional to the cells mass, you get the same (=poor) cycle life and slightly more saggy discharge curve.

When you think about it, the 18650 form factor is better for small applications and high discharge rates : more cooling surface and less distance between battery poles and internal windings. And the ability to increment your packs in smaller steps for better fit.

 

[Pajda]

Finally I got one tester channel free and start with cycle life test. It will take at least one month to get some initial results. But I have no problem with nominal capacity. On my calibrated 4-wire tester I got this results:

Nominal capacity test*: capacity: 4.795 mAh, energy: 17.400 Wh
My cycle life test 100%DoD**: capacity: 4.766 mAh, energy: 16.590 Wh

*0.3C charge(1440 mA) to 4.2V @ 100mAh cut-off -- 10min rest time -- 0.2C discharge (960 mA) to 2.5V
**0.5C charge(2350 mA) to 4.2V @ 100mAh cut-off -- 10min rest time -- 1C discharge (4700 mA) to 2.5V

 

[coinmaster]

So did we conclude these cells are a rip-off? Even if they come out at just 4000ma that is still better than anything else on the market by weight is it not?

 

[Pajda]

This Samsung 21700 format cell have very problematic cycle life when compared with most of 18650 cells. There are many cells in 18650 format on the market which have better performance, better cycle life, better energy density and they are also significantly cheaper. So after my testing I do not find this cell interesting.

 

[ElectricGod]

This Samsung 21700 format cell have very problematic cycle life when compared with most of 18650 cells. There are many cells in 18650 format on the market which have better performance, better cycle life, better energy density and they are also significantly cheaper. So after my testing I do not find this cell interesting.

I agree. I have 40 of them just sitting in a bag. Not the best purchase I've ever made. Oh well!

 

[ElectricGod]

I recently got an Icharger 4010 DUO and an ISDT D2. I have not used these cells at all except in bench tests. They have been sitting in a box hovering around 3.5 volts for a long time. I am charging and discharging them on the 4010 and the D2. I was curious to see if I got significantly different results than before on my D100.

Ir is usually in the 35-50 mOhm range.
Capacity is typically 3500mah.

I should make one note in my capacity testing. I'm discharging from 4.1 to 3.1 volts. I expect a little less capacity than I'd get if I went down to 2.5 volts. However, in all honesty, you would be stupid to bother with 2.5 volts in any real application.

I think it's time to find some more 21700 cells and do more testing. This early batch of INR21700-48BG cells were disappointing.

 

[Pajda]

I have some new info about cycle life tests for 48G.

Under my first standardized test 0.5C-1C 100% DoD settings the results should be marked as "disaster / non usable" I have run this setting on another cell sample with similar result. But then I lower down the charging current to 0.3C (manufacturer recomended nominal charging current) and now running the setting 0.3C-1C 100% DoD. The results are now much better and I give the mark" usable", but still there are cells in 21700 and 18650 format with better cycle life even with higher energy density.

Under my second standardized test 0.5C-1C 50% DoD (from 100% to 50% SOC) this 48G perform very well and is fully comparable to the best 18650 and 21700 cells on the market.

Short conclusion: This cell is usable if you plan to use slow charging /regen. But there are much better cells with lower price on the market. LG INR21700 M50 is better choice in all aspects including the price.

 

[ElectricGod]

IMHO, just don't buy this Samsung cell...it's not that great.