Samsung 40T and other cells like it?

Batteries, Chargers, and Battery Management Systems.
rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Samsung 40T and other cells like it?

Post by rg12 » Oct 05 2019 11:23am

Got this one brand new 40T cell by accident and it reads 7 milliohm internal resistance for a 4Ah cell that is not much bigger than a 18650 (a 21700).
Until now I didn't know about it and am looking for an 8C that has better power density than the 25R and this one seems to be it only I read that it can go to 35A but it heats up.
Anyone have any experience with it?
How does it compare to 25R?
Any other cells with similar specs?

john61ct   100 GW

100 GW
Posts: 5404
Joined: Dec 18 2018 2:06pm

Re: Samsung 40T and other cells like it?

Post by john61ct » Oct 05 2019 2:15pm

rg12 wrote:Got this one brand new 40T cell by accident and it reads 7 milliohm internal resistance for a 4Ah cell that is not much bigger than a 18650 (a 21700).
Until now I didn't know about it and am looking for an 8C that has better power density than the 25R and this one seems to be it only I read that it can go to 35A but it heats up.
Anyone have any experience with it?
How does it compare to 25R?
Any other cells with similar specs?
Yes great power discharge rates.

But not great lifespan. 30Q is worse though, per Pajda 40T gets 1000 cycles at 100% DoD with 63% of initial capacity left.

50E is **much** lower power discharge, maximum continuous discharge 2C (9800mA) and 3C (14700mA) not for continuous discharge,

but if that is not critical, can build higher Ah capacity, then with the same test 1000 cycles leaves 85% capacity, may last twice as long!

Note 2.5A charge rate per string is max for longevity.

Also check out LG INR21700H40 
http://queenbattery.com.cn/our-products ... l-37v.html

User avatar
flippy   1 MW

1 MW
Posts: 1681
Joined: Aug 12 2015 3:07pm

Re: Samsung 40T and other cells like it?

Post by flippy » Oct 05 2019 5:20pm

i got a couple here and they suck basically.
when they are new they are awesome and great. but once some wear is put on the cell it goes south fast and bad.

in no way should these cells be used for anyting you want to last.
sidenote: this base concept is valid for most high capacity cells.
Lithium beats liquid dinosaurs.

Pajda   1 kW

1 kW
Posts: 344
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Samsung 40T and other cells like it?

Post by Pajda » Oct 06 2019 2:27am

rg12 wrote:
Oct 05 2019 11:23am
Got this one brand new 40T cell by accident and it reads 7 milliohm internal resistance for a 4Ah cell that is not much bigger than a 18650 (a 21700).
Until now I didn't know about it and am looking for an 8C that has better power density than the 25R and this one seems to be it only I read that it can go to 35A but it heats up.
Anyone have any experience with it?
How does it compare to 25R?
Any other cells with similar specs?
Look at this excellent table of DC/ACIR made by docware https://i.imgur.com/FRRahFf.jpg you can see that Samsung 40T have lower internal resistance than 25R. In 40T datasheet they stated that it should do 35A continuous under 80°C. Yes 40T is losing its capacity fast, paticularly in first ca 100 cycles, but the more important DCIR grow in HP appplication is not bad at all. I hope I will test soon LG H40, I personally expect slightly higher DCIR but significantly better overall operational life, but it must be tested for sure.

rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 06 2019 8:19am

Pajda wrote:
Oct 06 2019 2:27am
rg12 wrote:
Oct 05 2019 11:23am
Got this one brand new 40T cell by accident and it reads 7 milliohm internal resistance for a 4Ah cell that is not much bigger than a 18650 (a 21700).
Until now I didn't know about it and am looking for an 8C that has better power density than the 25R and this one seems to be it only I read that it can go to 35A but it heats up.
Anyone have any experience with it?
How does it compare to 25R?
Any other cells with similar specs?
Look at this excellent table of DC/ACIR made by docware https://i.imgur.com/FRRahFf.jpg you can see that Samsung 40T have lower internal resistance than 25R. In 40T datasheet they stated that it should do 35A continuous under 80°C. Yes 40T is losing its capacity fast, paticularly in first ca 100 cycles, but the more important DCIR grow in HP appplication is not bad at all. I hope I will test soon LG H40, I personally expect slightly higher DCIR but significantly better overall operational life, but it must be tested for sure.
I don't understand the chart, what are the other IR numbers on the other columns?

I don't get it, how come 30Q and 25R have the same IR but one is rated for 8C and one for 5C?

Also, how come you say that 40T is good according to this chart while it goes bad really fast?

Any tested brand 21700 cells that have good lifespan and performance?
I need a good one for high capacity and high discharge rate and also one for highest capacity possible with normal discharge rate.
and what was that about high capacity cells that don't last long/lose capacity fast?

john61ct   100 GW

100 GW
Posts: 5404
Joined: Dec 18 2018 2:06pm

Re: Samsung 40T and other cells like it?

Post by john61ct » Oct 06 2019 9:51am


rg12 wrote:I don't understand the chart, what are the other IR numbers on the other columns?

I don't get it, how come 30Q and 25R have the same IR but one is rated for 8C and one for 5C?

Also, how come you say that 40T is good according to this chart while it goes bad really fast?

Any tested brand 21700 cells that have good lifespan and performance?
I need a good one for high capacity and high discharge rate and also one for highest capacity possible with normal discharge rate.
and what was that about high capacity cells that don't last long/lose capacity fast?
Tuning chemistry for high capacity reduces the discharge rate, you can't have both.

Longevity is independent of both, has more to do with usage patterns, repeated high rate discharging you will need to replace the pack more frequently anyway.

Focus on which attribute you **really** need, everything is a compromise, you can't have it all.

Pajda   1 kW

1 kW
Posts: 344
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Samsung 40T and other cells like it?

Post by Pajda » Oct 06 2019 2:53pm

rg12 wrote:
Oct 05 2019 11:23am
I don't understand the chart, what are the other IR numbers on the other columns?
DCIR IEC is a standardized "two stage method" of measuring DCIR, which requires slightly better equipment to perform, DCIR 10s is "single stage" method, which is very easy to perform, you have to just measure the voltage in 10th second of applying defined load current. Measuring of DCIR is tricky, because there is no "true" absolute DCIR value usable for all cases, so the common approach is to choose method, which is close to your mode of operation and then just compare the cells side by side. You can see in this table, that those two methods are mostly only shifted.
rg12 wrote:
Oct 05 2019 11:23am
I don't get it, how come 30Q and 25R have the same IR but one is rated for 8C and one for 5C?
My opinion is that Samsung do this because cycle life of 30Q, ale least from initial production batch sucks.
rg12 wrote:
Oct 05 2019 11:23am
Also, how come you say that 40T is good according to this chart while it goes bad really fast?
If you perform "a lot" of your own measurements, you find out that there are to little generalization rules to follow when evaluating li-ion cells. In fact most of informations about cell life which you find on internet are only "old wives' advice" based on rules which actually did not works in significant amount of cases :?

Many people forget when evaluating cell EOL, that there are two major conditions. Capacity fade and DCIR rise, where the DCIR rise is important particularly in applications with more than 0.2C (rated capacity) discharge load. Also there is of course no general rule about dependency between this two parameters. So you can find many cells with high capacity fade but with relative stable DCIR and also cells with relative low capacity drop but significant DCIR rise, which make some cell almost unusable just at ca 85% of remaining capacity.

By the way he asked for 8C discharge cell and 40T can do this, I am thinking that nobody sensible will expect that 40T or similar cell design can do much more than 250 cycles under this continuous load or the discharge end cell temperature will be significantly lower than 80°C.

User avatar
flippy   1 MW

1 MW
Posts: 1681
Joined: Aug 12 2015 3:07pm

Re: Samsung 40T and other cells like it?

Post by flippy » Oct 06 2019 4:34pm

Pajda wrote:
Oct 06 2019 2:53pm
By the way he asked for 8C discharge cell and 40T can do this, I am thinking that nobody sensible will expect that 40T or similar cell design can do much more than 250 cycles under this continuous load or the discharge end cell temperature will be significantly lower than 80°C.
from my tests those cells at 8C will be shot after <100 cycles and completly dead shortly after 100. at least if the battery does not melt beforehand.
Lithium beats liquid dinosaurs.

Pajda   1 kW

1 kW
Posts: 344
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Samsung 40T and other cells like it?

Post by Pajda » Oct 07 2019 1:19am

flippy wrote:
Oct 06 2019 4:34pm
Pajda wrote:
Oct 06 2019 2:53pm
By the way he asked for 8C discharge cell and 40T can do this, I am thinking that nobody sensible will expect that 40T or similar cell design can do much more than 250 cycles under this continuous load or the discharge end cell temperature will be significantly lower than 80°C.
from my tests those cells at 8C will be shot after <100 cycles and completly dead shortly after 100. at least if the battery does not melt beforehand.
Interesting, I run tests only up to 3C continuous discharge and 40T perform relatively well (almost same capacity fade as under 1C continuous with usable DCIR rise). Information about 250 cycles at 35A (8C) continuous comes directly from 40T datasheet.

User avatar
flippy   1 MW

1 MW
Posts: 1681
Joined: Aug 12 2015 3:07pm

Re: Samsung 40T and other cells like it?

Post by flippy » Oct 07 2019 3:30am

i did a test with 100 cycles and after 43 cycles the tester faulted out due to overheating the cell. i have set the upper limit to 75c in order to prevent cells from popping its top.
after those 43 cycles it already lost quite a bit of capacity and IR was also higher.

i think they tested these cells with active cooling, not in a holder or something so free air convection was minimal in my case but that should still not result in borderline melting the cell holder imho.

i am running a test on a single cell under much more reasonable loads and see what happens, but so far i am not impressed considering the cost of the cell.
Lithium beats liquid dinosaurs.

rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 07 2019 4:13am

flippy wrote:
Oct 07 2019 3:30am
i did a test with 100 cycles and after 43 cycles the tester faulted out due to overheating the cell. i have set the upper limit to 75c in order to prevent cells from popping its top.
after those 43 cycles it already lost quite a bit of capacity and IR was also higher.

i think they tested these cells with active cooling, not in a holder or something so free air convection was minimal in my case but that should still not result in borderline melting the cell holder imho.

i am running a test on a single cell under much more reasonable loads and see what happens, but so far i am not impressed considering the cost of the cell.
Those are insane temps, how can Samsung recommend you to operate a cell near 80c?
I use mostly 25R and had a few customers who ran the pack at 5.5C and after 10 month the IR went so high in all cells and the pack was junk and unusable so my rule of thumb is no more than 50% use of C rate for all cells.

You personally have alot of experience with actual cell performance...
Do you know of any 21700 cell that has exceptionally high capacity while having a high discharge rate that has a good lifespan if used only with 50% of it's advertised discharge rate?

User avatar
flippy   1 MW

1 MW
Posts: 1681
Joined: Aug 12 2015 3:07pm

Re: Samsung 40T and other cells like it?

Post by flippy » Oct 07 2019 5:18am

the tesla ludicrous mode (or supercharging) jacks up the temp to 45c and keeps it there as the critical cell temp is 55 and the higher temp lowers the IR. but due to the massive cooling/heating system it can maintain that without much issue. the model 3 is better then the S/X in this regard as it has a better battery thermal design.
tesla starts kneecapping the power at 55c and will do anything it needs to to keep it below that. so i dont know that samsung is thinking with the T40 datasheet numbers. running those cells at those temps is just stupid.
Lithium beats liquid dinosaurs.

User avatar
pwd   1 kW

1 kW
Posts: 372
Joined: Oct 11 2011 12:01pm
Location: Ontario, Canada

Re: Samsung 40T and other cells like it?

Post by pwd » Oct 07 2019 8:51am

Pajda wrote:
Oct 06 2019 2:27am
... I hope I will test soon LG H40
Excellent, I'm looking forward to that. It seems that Queen Battery is the only place to stock that cell.
Commuter/Street Build viewtopic.php?f=6&t=98286:
Rocky Mountain Element
Magic Pie 4 front + Leafmotor 1500 rear w/ WCEC 18fet
Samsung 40T 14S7P

Offroad Build viewtopic.php?f=6&t=102195:
2019 Rocky Mountain Blizzard 20
Cyclone "3kw" 13T motor to 36T chain ring then 32T chain ring to 11T-46T Cassette
em3ev 14S5P 30Q or 14S6P 25R pack

john61ct   100 GW

100 GW
Posts: 5404
Joined: Dec 18 2018 2:06pm

Re: Samsung 40T and other cells like it?

Post by john61ct » Oct 07 2019 9:33am


rg12 wrote:Do you know of any 21700 cell that has exceptionally high capacity while having a high discharge rate
Remember, those two goals are mutually exclusive.

Decide what your actual current will be, peak for 10-sec, 5-min max and continuous.

If that aspect is most important, forget per-cell capacity, ideally size the whole bank to a higher Ah capacity to bring down the C-rate.

And forget vendor max C-ratings, pure fiction, work off these testers' real-life data taking IR / temps into account.

That should narrow things down, but the longevity part is not going to be definitive.



rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 08 2019 8:58am

Well, if we are on the topic...
What is the highest capacity I can find in Li Ion cylindrical? (above 21700 of course)

Pajda   1 kW

1 kW
Posts: 344
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Samsung 40T and other cells like it?

Post by Pajda » Oct 08 2019 1:05pm

rg12 wrote:
Oct 08 2019 8:58am
Well, if we are on the topic...
What is the highest capacity I can find in Li Ion cylindrical? (above 21700 of course)
Sorry, I do not understand the language of your tribe. You mean what is the biggest cylindrical cell on the market? If you are looking for best absolute energy density in cylindrical cell, it is still in 18650 and 21700 format, smaller or bigger fomats have lower energy density.

docware   1 kW

1 kW
Posts: 317
Joined: Sep 16 2018 5:36am

Re: Samsung 40T and other cells like it?

Post by docware » Oct 08 2019 2:26pm

rg12 wrote:
Oct 08 2019 8:58am
Well, if we are on the topic...
What is the highest capacity I can find in Li Ion cylindrical? (above 21700 of course)
The problem with your „Monster bike“ is that your 25R 24s 7p battery is not properly designed to your 5 kW motor. For your expected 80 – 100 A you need much higher capacity, much bigger battery pack dimensions, 30 - 40 Ah. Forget about nonsense like 8C or 5C.

viewtopic.php?f=6&t=100788&p=1474589#p1474589
bike1.jpg
bike1.jpg (353.91 KiB) Viewed 2306 times

rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 08 2019 4:13pm

Pajda wrote:
Oct 08 2019 1:05pm
rg12 wrote:
Oct 08 2019 8:58am
Well, if we are on the topic...
What is the highest capacity I can find in Li Ion cylindrical? (above 21700 of course)
Sorry, I do not understand the language of your tribe. You mean what is the biggest cylindrical cell on the market? If you are looking for best absolute energy density in cylindrical cell, it is still in 18650 and 21700 format, smaller or bigger fomats have lower energy density.
Well, I know about 32650 and such and know that there are more fatty sizes to lithium like it and was wandering which one has the best energy density, yeah.
So far it seems that the 21700 is the winner with a 5000mAh capacity which 32650 which is a huge cell has only about 6000mAh max.

docware wrote:
Oct 08 2019 2:26pm
rg12 wrote:
Oct 08 2019 8:58am
Well, if we are on the topic...
What is the highest capacity I can find in Li Ion cylindrical? (above 21700 of course)
The problem with your „Monster bike“ is that your 25R 24s 7p battery is not properly designed to your 5 kW motor. For your expected 80 – 100 A you need much higher capacity, much bigger battery pack dimensions, 30 - 40 Ah. Forget about nonsense like 8C or 5C.

viewtopic.php?f=6&t=100788&p=1474589#p1474589
bike1.jpg
I'm not looking for a cell for my own bike.
I ride my bike about once every few month so I don't mind beating my pack for one cycle with only a few bursts of high power here and there.
My pack is 17.5Ah 25R and I'm pulling about 7C from it in bursts once every few month so fudge it.

john61ct   100 GW

100 GW
Posts: 5404
Joined: Dec 18 2018 2:06pm

Re: Samsung 40T and other cells like it?

Post by john61ct » Oct 08 2019 5:39pm

For maximum energy density - and power density also, most likely, look at RC hobby LiPo packs.

6S units

Rhino is a well respected vendor.
https://hobbyking.com/en_us/rhino-5000m ... -xt90.html

Turnigy, Tattu, not as much

Forget vendor C ratings, but with these 5-8C in real life is at the low end.

Don't expect great longevity (ab)using them that way though, going to a higher total Ah capacity to keep discharge below 4C, and stopping well above 10% SoC you'll get better value from their high $/Ah

rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 08 2019 5:50pm

john61ct wrote:
Oct 08 2019 5:39pm
For maximum energy density - and power density also, most likely, look at RC hobby LiPo packs.

6S units

Rhino is a well respected vendor.
https://hobbyking.com/en_us/rhino-5000m ... -xt90.html

Turnigy, Tattu, not as much

Forget vendor C ratings, but with these 5-8C in real life is at the low end.

Don't expect great longevity (ab)using them that way though, going to a higher total Ah capacity to keep discharge below 4C, and stopping well above 10% SoC you'll get better value from their high $/Ah
Not interested in lipos.
Of course non-cylindrical cells have more energy density but they are not as safe, especially lipos

Shunt   100 mW

100 mW
Posts: 45
Joined: Aug 06 2019 9:39pm

Re: Samsung 40T and other cells like it?

Post by Shunt » Oct 08 2019 6:42pm

The Queen Battery QB26800 doesn't quite hit the current rating to compete with the high discharge Samsungs , but it has a whopping capacity per cell . I haven't used these personally to comment on longevity , but they apparently test well in reviews.

docware   1 kW

1 kW
Posts: 317
Joined: Sep 16 2018 5:36am

Re: Samsung 40T and other cells like it?

Post by docware » Oct 09 2019 1:15am

Well rg12, then 21700 format may be interesting for you. Samsung 50E for high energy application and Samsung 40T for high power application. See informative graphs for single cell warming in the BF-2A fixture. Here is cell cooled by ambient air and also by BF-2A itself. Of course, in the battery pack is warming different.
50E warming at 5A.jpg
50E warming at 5A.jpg (74.01 KiB) Viewed 2261 times
40T warming at 10A.jpg
40T warming at 10A.jpg (115.02 KiB) Viewed 2261 times
1.jpg
1.jpg (116.25 KiB) Viewed 2261 times

rg12   100 kW

100 kW
Posts: 1278
Joined: Jul 26 2014 4:22pm

Re: Samsung 40T and other cells like it?

Post by rg12 » Oct 09 2019 6:59am

docware wrote:
Oct 09 2019 1:15am
Well rg12, then 21700 format may be interesting for you. Samsung 50E for high energy application and Samsung 40T for high power application. See informative graphs for single cell warming in the BF-2A fixture. Here is cell cooled by ambient air and also by BF-2A itself. Of course, in the battery pack is warming different.

50E warming at 5A.jpg

40T warming at 10A.jpg

1.jpg
Thats very interesting!
Is that from the datasheet?
I would like to know the temps for about double these currents for about 1 minute constant

docware   1 kW

1 kW
Posts: 317
Joined: Sep 16 2018 5:36am

Re: Samsung 40T and other cells like it?

Post by docware » Oct 09 2019 10:47am

rg12 wrote:
Oct 09 2019 6:59am
Thats very interesting!
Is that from the datasheet?
I would like to know the temps for about double these currents for about 1 minute constant
It´s my measurement.
1 minute is too short time to see the temperature change on the surface of the cell. Can you describe intention of such measurement more extensively ?

john61ct   100 GW

100 GW
Posts: 5404
Joined: Dec 18 2018 2:06pm

Re: Samsung 40T and other cells like it?

Post by john61ct » Oct 09 2019 12:21pm

Pajda wrote:
Oct 06 2019 2:53pm
Many people forget when evaluating cell EOL, that there are two major conditions. Capacity fade and DCIR rise, where the DCIR rise is important particularly in applications with more than 0.2C (rated capacity) discharge load. Also there is of course no general rule about dependency between this two parameters. So you can find many cells with high capacity fade but with relative stable DCIR and also cells with relative low capacity drop but significant DCIR rise, which make some cell almost unusable just at ca 85% of remaining capacity.
Seems contradicted by Frank over here https://www.rcgroups.com/forums/showthr ... p=42911965

> Loss of capacity and increase of internal resistance are connected: "Active" lithium reacts, forms deposits at the SEI layer and become "inactive". SEI layer thinkness increases and thus resistance.

Post Reply