High drain 21700 - Samsung 30T discharge test

thunderheart · Jul 9, 2018

Original article: https://www.thunderheartreviews.com/2018/08/high-drain-21700-li-ion-samsung-30t.html

I've tested the Samsung INR21700-30T (35A) at currents up to 20A (limited by my battery tester).

As always the cell was bought from 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 followed all the prescriptions of the IEC61960-2003 standard concerning battery's capacity measurement. Before each discharging cycle each battery was charged at standard current mentioned in its datasheet to charge end voltage (4.2V) (cut-off at 0.1A, which is the lowest supported by EBC-A20). Before each discharging or charging i've held a 1-1.5hrs pause. The environment temperature was about 25°C.

Samsung INR21700-30T

The cell is marked as INR21700-30T SAMSUNG SDI 2HC2

The main specs from its datasheet:
Minimum capacity: 3000mAh
Nominal voltage: 3.6V
Standard charge current: 1.5A
Max. charge current: 4A
Charge end voltage: 4.2V
Charge cut-off current: 100mA
Max. discharge current: 35A
Discharge cut-off voltage: 2.5V
Max weight: 69g

The measured weight is 68.22g

Test results:

The results are really impressive! At 0.2C it did 3155mAh/11.66Wh and it's capacity was above 3000mAh even at 20A discharge rate! Pay attention to the curves at 10A and 15A - they are almost identical. On all 5 curves there are no voltage drops or sudden peaks and the 30T seems to be rock solid like Steven Seagal - it does the most difficult job without sweating

I hope this test was interesting to read/watch and i'll be happy if it was useful for someone

Here is the video version of this review with size comparison between 18650 and 21700:
[youtube]BAPtWjBBah0[/youtube]

P.S. this is my YouTube channel: www.youtube.com/thunderheartreviews It will be highly appreciated if you subscribe!

pwd · Jul 10, 2018

Thanks again for the excellent test. This is a very interesting cell, if I was to go with a 21700 cell; this would be it.

thunderheart · Jul 10, 2018

Thanks pwd! It's a pity i couldn't discharge it at 35A

madin88 · Jul 11, 2018

Yeah the Samsung 30T is awesome in terms of power. It is much closer to LiPos as any other cells. Thank you for sharing the test.

I see you bought the 30T from Qeen Battery. How much was it?
Does anybody know of other sources where they can be bought (in larger quantities not just a few for e-cigarette usage)?

Queen battery also does list some other very interesting high drain cells i never heard from:

Sanyo NCR20700C 3500mAh 35A
Sony US21700VTC6A 4200mAh 30A
Samsung INR21700-40T 4000mAh 30A

If thats no fake than the future looks promising 8)

thunderheart · Jul 11, 2018

I see you bought the 30T from Qeen Battery. How much was it?

It was $5 or 7, i don't remember the exact price.
It's out of stock now.

If thats no fake than the future looks promising

i think 21700 will kill 18650. The future will show))

Allex · Sep 4, 2018

Samsung 30T
Sanyo NCR20700C 3500mAh 35A
Samsung INR21700-40T 4000mAh 30A

All the models above are the latest 20700/21700 batteries, which most of them are not bulk manufactured yet or so they say.

Hillhater · Sep 4, 2018

Did you recheck those 10 & 15A discharge results ?
Being so near identical is unprecidented and practically impossible !....considering the other results .

Hillhater · Sep 4, 2018

For interest / comparason, these are the 30-40A discharge results from the Ecig forum..
They rate this cell as the best available high drain cylindrical cell in this size range.

Matador · Sep 4, 2018

Thanks !!!

This Samsumg 30T is "The Sony VTC4" of 21700 cells.
Low resistance, high current, and the capacity/volume seems proportional...

Interesting cell indeed !

BCTECH · Sep 5, 2018

next question is where do we get these cells and in what price.
is queen battery the only place selling them?

thunderheart · Sep 5, 2018

Hillhater said:
Did you recheck those 10 & 15A discharge results ?
Being so near identical is unprecidented and practically impossible !....considering the other results .

Hillhater said:
For interest / comparason, these are the 30-40A discharge results from the Ecig forum..
They rate this cell as the best available high drain cylindrical cell in this size range.

All tests i do at least twice and more often three times to be sure in results. My results never depend on someone else's results found on the net. BTW i've tested LG HB4 vs HB6 and found that HB6 is better in all tests but according to another reviewer HB4 is better. What can i say? I trust what i've got in my hands rather than someone else's results.

P.S. When i was comparing M50 with 48G i re-tested another 48G cell i got and the results were the same as when i was testing 48G before (4-9mAh difference). So i have no reason to not trust my own tests.

thunderheart · Sep 5, 2018

BCTECH said:
next question is where do we get these cells and in what price.
is queen battery the only place selling them?

As far as i know they are out of stock, so you can't get them from Queen Battery. Drop them an e-mail to check availability.

Hillhater · Sep 5, 2018

thunderheart said:
...
All tests i do at least twice and more often three times to be sure in results. My results never depend on someone else's results found on the net. .......

Its not a matter of comparing with other test results,....it a matter of simple electrical physics.
Assuming all other factors remain constant , more current would result in more voltage sag....just as it did with all your other discharge tests.
If those 10/15 A results are truely accurate, then there is something very, very, unique happening ! :shock:

But, IF you do care to compare your results to those from e-cig, or other sources, you may notice a very significant difference for all those discharge tests.
Your cell appears to display voltage sag much more (~0.3v). for any given load, than the cell tested by e-cig or lygte.. ??

So either you had a bad cell, or your test results are inaccurate.

PS.. This is not a critisism, but hopefully a helpful observation....you will also notice a similar voltage discrepancy in your VTC4 results , compared to others linked in that thread ?

thunderheart · Sep 6, 2018

The higher voltage sag in the beginning could be result of 1-1.5hr pause which i hold in respect to the IEC61960-2003 standard. I don't know if the others keep that pause or consider it as a loss of time but the standard has clear instructions how to test a battery.
In Russian segment of Internet there are also some people who test batteries - some of them start discharging right after charging is finished. It results in higher initial voltage and higher capacity but it's not the way a test should be done. There are clear rules and a tester should follow them even if he has to wait an hour before discharging.

I'm open to any argumented critisism or debate, so everything is OK

Allex · Sep 6, 2018

thunderheart said:
some of them start discharging right after charging is finished. It results in higher initial voltage and higher capacity but it's not the way a test should be done. There are clear rules and a tester should follow them even if he has to wait an hour before discharging.

This is true, something to keep in mind!

madin88 · Sep 6, 2018

thunderheart said:
As always the cell was bought from 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.

The tester does 4-wire measuring, but it looks like it does it only towards the alligator clip, and from there on it will be just 2-wire.
It would better (more accurate) if you convert it to 4 alligator clips. Then you have two for the current and two for the voltage measurement.
There probably will not be much difference, but as it is now there definitely will be a small voltage drop over the alligator clip!

I think such cell holder from an IR-meter would be a nice upgrade:

thunderheart said:
The higher voltage sag in the beginning could be result of 1-1.5hr pause which i hold in respect to the IEC61960-2003 standard. I don't know if the others keep that pause or consider it as a loss of time but the standard has clear instructions how to test a battery.

Yes that will have an effect.
No pauses would mean the cell cannot cool down after discharge, and the higher the temperature the lower the internal resistance which leads to less voltage sag.

Pajda · Sep 6, 2018

madin88 said:
The tester does 4-wire measuring, but it looks like it does it only towards the alligator clip, and from there on it will be just 2-wire.
It would better (more accurate) if you convert it to 4 alligator clips. Then you have two for the current and two for the voltage measurement.
There probably will not be much difference, but as it is now there definitely will be a small voltage drop over the alligator clip!

I agree with this point, I will also recommend you to use four alligator clips. In fact this can be the main reason for the lower measured voltage. The one hour rest time does not affect this significant voltage drop difference under load by my guess.

The other important isuue is that your tester have minimal charging cutt-off current 100 mA, where the most cells needs 50 mA cut-off current for measure of nominal capacity. But this problem also does not affect the significant voltage drop, only the measured nominal capacity.

thunderheart · Sep 6, 2018

madin88 said:
if you convert it to 4 alligator clips. Then you have two for the current and two for the voltage measurement.
There probably will not be much difference, but as it is now there definitely will be a small voltage drop over the alligator clip!

Yes, it will become more accurate but i think the accuracy will rise by something like 0.001% which is miserable advantage to deal with. I'm testing not only 18650/21700 cells but also Lipos and cased prismatic cells using self-made adapters. If i double the clips it would complicate using the tester with adapters, so i'll have to complicate my adapters too.

madin88 said:
I think such cell holder from an IR-meter would be a nice upgrade:

It would be a nice downgrade as it's contact square is much less than my holder's especially of the version 3.0 of my holder which you'll see in my next review in a couple of days

I was looking at that holder a couple of years ago before but decided to make my own holder with larger contacts made of pure copper.

Pajda said:
The other important isuue is that your tester have minimal charging cut-off current 100 mA, where the most cells needs 50 mA cut-off current for measure of nominal capacity. But this problem also does not affect the significant voltage drop, only the measured nominal capacity.

It could affect the voltage drop in the beginning because the "more charged" cell will have higher voltage after 1hr pause but, as you said, the difference should not be significant. Whatever, my tester's minimum is 100mA and i have to deal with it...

madin88 · Sep 6, 2018

thunderheart said:
It would be a nice downgrade as it's contact square is much less than my holder's especially of the version 3.0 of my holder which you'll see in my next review in a couple of days I was looking at that holder a couple of years ago before but decided to make my own holder with larger contacts made of pure copper.

Yes the contact area is lower as it is on your self made holder, but the important thing is there is a pin inside the contact for measuring the voltage exactly at the cell.
Did you miss that?
The voltage drop in the current supply line then does not matter anymore!

If i double the clips it would complicate using the tester with adapters, so i'll have to complicate my adapters too.

A bit more complicated, yes, but wouldn't it be worth it as trade off for having more accurate results?
IMO not be a big deal to cut off the small wires and solder two new alligator clips.

Looking forward to see the new holder.

thunderheart · Sep 6, 2018

madin88 said:
A bit more complicated, yes, but wouldn't it be worth it as trade off for having more accurate results?
IMO not be a big deal to cut off the small wires and solder two new alligator clips.

Looking forward to see the new holder.

Not a big deal at all, it's 10 minutes of work but is it worth the 0.001% more accurate results i'd have? For information: my tester's accuracy is somewhere around 1%

Hillhater · Sep 6, 2018

thunderheart said:
....
..... it's 10 minutes of work but is it worth the 0.001% more accurate results i'd have? For information: my tester's accuracy is somewhere around 1%

But there appears to be a 5-10% difference between your voltage results and others, so it should be worthwhile to check that possibility.
......From the lygte test proceedure....
..Note the type of battery clamp he uses. :wink:

.......I want a very good connection to the battery and also separate wires for measuring voltage, i.e. the voltage is measured at the battery, not at the load or the power supply. This way I can get rid of any voltage drop in wires and connections. Before I have used a clamp for battery connection and this has worked well. I have connected 3 set of wires to the clamp: Load, power supply and measurement wires. The connection to the battery is a piece of copper.
The actual voltage drop from the battery to the measurement points is below 0.01 volt, even at 5 ampere load.

.........
After the charge I let the battery rest for 1 hour, where the voltage again will drop a bit, before I start a discharge test.
It can also be seen on the graph that my power supply is slightly below 4.2 volt, it is 4.196 volt.

. https://lygte-info.dk/info/Batteries2012Info%20UK.html

thunderheart · Sep 6, 2018

I've already explained my testing procedure (IEC61960-2003 standard) and i don't think i should change it to get closer to someone's results. Also don't forget that there are no 2 absolutely identical cells. To understand that compare LG HB4 vs HB6 on lygte.dk with my results of the same models.

Concerning his holder - it has much smaller contact area with battery than mine, the terminals on the wires are located in about 1cm from the battery (each) and they are made of a metal other than copper. What's the advantage over my holder i should note?

Hillhater · Sep 6, 2018

thunderheart said:
I've already explained my testing procedure (IEC61960-2003 standard) and i don't think i should change it to get closer to someone's results. Also don't forget that there are no 2 absolutely identical cells. To understand that compare LG HB4 vs HB6 on lygte.dk with my results of the same models.

Concerning his holder - it has much smaller contact area with battery than mine, the terminals on the wires are located in about 1cm from the battery (each) and they are made of a metal other than copper. What's the advantage over my holder i should note?

Again, the difference between your HB4, HB6, results and those on lygte ..is a consistent 0.3v (approx) lower result at any given SOC and load. (Compare with 20A load, at 0.5 Ah and 1.0Ah discharge points )
That is a 10% difference !
This should be ringing some alarm bells in your mind by now.
Since your voltage readings are lower, it would suggest you have a higher resistance in your voltage measurement circuit (about 0.015 ohm).
Have you double checked cell voltages, using an accurate, calibrated quality dmm direct onto the cell caps ?
Its not the size of those contacts that is important, its their resistance, and a smaller contact point would mean a higher contact pressure, = better contact. ?
Tightly clamped terminals are probably less resistance than crock clips.

liveforphysics · Sep 6, 2018

Your 10amp and 15amp data sets have a critical error and are not possible, the waiting between datasets is improbable to be even a percent or two difference, the lower impedance the cell the shorter the recovery interval it would need to achieve charge distribution equilibrium in the active material.

If you want to get clean data, I would use a mechanically separate wire whisker/needle probe for the cell voltage value. Drive the current from a contact point at least ~2-3mm away from where your voltage measurement happens.

thunderheart · Sep 6, 2018

Hillhater are you trolling me???
Look at these graphs

What do you see? HB4 is better than HB6. And the picture is the same at other discharge rates

Now look at my results:

the picture is completely different - HB6 is better than HB4 at all rates.

Now tell me about what consistent 0.3V or 10% difference you're talking about? If my holder is a "bad" one with higher resistance then i'd get the same picture (higher HB4 results) but with a bit lower capacity and bigger voltage drop. BUT my results show that HB6 is better. Once again, if my holder's resistance would be higher, the only difference with lygte's result would be the lower capacity and bigger sag. Verdict: the difference in results is not caused by the holder.

liveforphysics said:
Your 10amp and 15amp data sets have a critical error and are not possible, the waiting between datasets is improbable to be even a percent or two difference

Really?))) And what's that critical error? And what you mean by "waiting between datasets"?

liveforphysics said:
If you want to get clean data, I would use a mechanically separate wire whisker/needle probe for the cell voltage value. Drive the current from a contact point at least ~2-3mm away from where your voltage measurement happens.

The data i get is more than enough clean for almost any usage. I don't aim to achieve 0.0000000001% accuracy - it's useless in case of batteries.

High drain 21700 - Samsung 30T discharge test

100 W

10 kW

100 W

10 MW

100 W

100 MW

100 TW

100 TW

100 kW

1 kW

100 W

100 W

100 TW

100 W

100 MW

10 MW

10 kW

100 W

10 MW

100 W

100 TW

100 W

100 TW

100 TW

100 W

Similar threads