Cycle life tests of High Energy density cylindrical cells

[From-A-To-B]

Random question. Why test 90% DoD? I daily charge my batteries to 80% or 90% capacity and then charge once in a while to 100% right before a long ride. I think I only discharge down to about 3V per cell. So, I guess I answered my own question.

Which aspect of the 100% DoD test do you think is more damaging to the cells leading to the differences you see between 100% DoD and 90% DoD test results? The charge to 4.2V or the discharge to 2.5V?

I did quite a bit of reading about this in the past, and cannot quickly find the most helpful sources. Some of those threads were here on ES, so it might be worth a search. Previously, I'd read that the top 10% and 10% are about equally as damaging. Thus, I keep my batteries between 90% and 10% SOC whenever I can help it. It does me little good to play around between 3.0 and 2.5 volts per cell, anyway, since I'm worried about getting home at that low level. My battery gets really soft and LVC is just around the corner when I near 3.0V per cell.

On the contrary, this article summarizes some testing that indicates that deep discharges are less impactful than charging to 4.2V per cell. Battery charging: Full versus Partial - PushEVs

At the end of the day, I just take it easy on the battery and know that I don't ride enough to see 1000 full cycles, anyway!

 

[Pajda]

Random question. Why test 90% DoD? I daily charge my batteries to 80% or 90% capacity and then charge once in a while to 100% right before a long ride. I think I only discharge down to about 3V per cell. So, I guess I answered my own question.

The general rule is that synthetic tests are one thing and real-life tests are another. The most difficult thing is just to accept this fact. After that, you just need to choose a standardized test and try to keep the conditions constant. The reason why 4.15-3.00V (about 90% of DoD) is that someone from ES community just asked for it. And the technical reason is that some modern cells can withstand 1000 cycles even at 100% DoD, and as you can see, 90% DoD cycling survived pretty much all samples. So going below 90% means that you won't see significant differences between the cells anymore.

Which aspect of the 100% DoD test do you think is more damaging to the cells leading to the differences you see between 100% DoD and 90% DoD test results? The charge to 4.2V or the discharge to 2.5V?

No general rule here either. I've seen several times a behaviour which does not make a sense. But again, discharging to 2.50V cut-off limit makes practical sense with HP cells at +5C discharge rate or at cold temperatures where the IR ensures that the cell is not fully discharged at this cut-off voltage.

 

[Pajda]

Also, why did you remove the Samsung 50S?

I don't like this cell? The problem I am aware of is the lack of a reasonable online database with graphical interface, where specific cells can be selected for comparison. So it is not only about 50S but many other cells data.

 

[tourist_trapper]

Small update in N38+F40 post. I have added some new informations in the Table. Added temperatures measured without forced air (I just turn off the thermal chamber after conditioning the cells to 25°C, so it acts as isolated box with 100l volume of air. There could be further confusion here , as modern best HE cells handle 3C continuous discharges just fine. No they don't.

Also added results of EVE INR18650 35V which seems to be also a decent cell. Cycle life tests started today.

After switching to the new version of the ES forum, I noticed a limit of 10 images per post. Unfortunately this now restricts me enough to move or add the new results to the first posts of the thread

Any chance you had time to complete the lifecycle tests for the 35v? (considering buying a bunch of reclaimed ones for projects and wondering if they're worth it)

 

[Kajman]

I don't like this cell? The problem I am aware of is the lack of a reasonable online database with graphical interface, where specific cells can be selected for comparison. So it is not only about 50S but many other cells data.

Will it be back? Very interested how is cycle life in such strong 5ah cell. Or any comment about it?

 

[Pajda]

Will it be back? Very interested how is cycle life in such strong 5ah cell. Or any comment about it?

The Samsung 50s by my standards today is no longer a champion in either the HP or HE category. But if you ask me, yes, samples of the 50s are on their way to me and I will measure it again.

 

[riba2233]

That would be nice, thanks! I have a hunch that their price might come down so they might be a good option for folks who need just something between hp and he.

 

[Souvlucky]

Certainly looking forward to the new Samsung 5.33Ah 21700 cell and its 15A/~3C pulse rating.
(The +1C fast-charging ability is also quite interesting—the M58/M58T and the M52V/M52VT being respectively rated at 0.5C and 0.7C max. continuous in their datasheets).

Might there be a new HE-midpower category brewing?

 

[Pajda]

Certainly looking forward to the new Samsung 5.33Ah 21700 cell and its 15A/~3C pulse rating.

Might there be a new HE-midpower category brewing?

I don't think so. The cell from samsung is supposed to have the designation INR21700-58E, i.e. the standard representative of their HE series, where the 5.33 Ah capacity is rather disappointing for today (competitors have 5.6 Ah). It seems to me that Samsung is currently lagging behind the competition, at least in the cylindrical cell category.

Listing the possibility of 1C charging in the datasheet, with a disclaimer against lifetime, means nothing at all.

 

[Souvlucky]

I don't think so. The cell from samsung is supposed to have the designation INR21700-58E, i.e. the standard representative of their HE series, where the 5.33 Ah capacity is rather disappointing for today (competitors have 5.6 Ah). It seems to me that Samsung is currently lagging behind the competition, at least in the cylindrical cell category.

Listing the possibility of 1C charging in the datasheet, with a disclaimer against lifetime, means nothing at all.

Quite – the INR21700-50E(1) did show relatively good cycle life in the +1C/100% DoD regime a couple years back.
It seems the latest INR21700-50G(7) was pretty solidin that regard as well. Of course it remains to be seen whether these advances roll-forward to newer cells, such as the INR21700-53G or the INR21700-58E.

About the INR21700-58E, a link to its technical datasheet found in the open may be useful.
(The cell is code-named CC5493F101 – nominal rating is 5.49Ah).
The cell's pricing should apparently be quite competitive as it is now nearing mass production for 2024/Q3.
One thing that is notable is the 3.64V nominal cell voltage.
Even at the typical 5.33Ah given rating (n.b: this is at -1C) the cell should have somewhere 19.4Wh~20Wh available in 100% DoD.

In any case, I have to agree with the statement that Samsung seems to lag behind the competition in the cylindrical category at least.
There have also been newer cells announced recently, such as the INR21700-50V and INR21700-50U.
As is often the case little to no new information has been made official, and we'll have to wait for these cells to appear in the open market to be tested and benchmarked.

 

[Pajda]

Among other things, a flood of interesting cells in the HE category are about to hit the market, especially in the 21700 format. It was about time, after all, as I was slowly running out of samples that I found interesting.

It's even led to the point where I've looked into a different category/technology and if all goes according to plan, I should have enough material to open a new thread here on ES this summer.

 

[tourist_trapper]

Among other things, a flood of interesting cells in the HE category are about to hit the market, especially in the 21700 format. It was about time, after all, as I was slowly running out of samples that I found interesting.

It's even led to the point where I've looked into a different category/technology and if all goes according to plan, I should have enough material to open a new thread here on ES this summer.

any idea on when they might hit the market?

 

[riba2233]

Any chance you would consider eve 50E to see how it stacks up to M50LT? They have pretty similar specs on paper but eve is very cheap for a 5ah 15a cell.

 

[Pajda]

Any chance you would consider eve 50E to see how it stacks up to M50LT? They have pretty similar specs on paper but eve is very cheap for a 5ah 15a cell.

Again, the biggest problem is that they are out of stock at Nkon (40P was supposed to arrive last week and they pushed it back a month, 50E even longer)

 

[riba2233]

Again, the biggest problem is that they are out of stock at Nkon (40P was supposed to arrive last week and they pushed it back a month, 50E even longer)

Ah ok, hopefully they get them back soon, really excited for them and thank you a lot!

 

[ediskrad]

Important information for those who will consider buying a Vapcell F60 for assembling large batteries. We assembled the battery in 16s14p format and at a constant current of 50 amperes, the battery overheated 5 times (reached the BMS cutoff of 60 degrees Celsius). The plant subsequently announced that it does not recommend assembling blocks of more than 40 elements from these elements. Quote "we suggest not to make 4000 and 6000mAh capacity for large battery pack , the limited cells for one pack should be less than 20 pcs ; if you want more cells to make pack ,please control 4S if more than 40 pcs , the current should not exceed 15Ah"

 

[harrisonpatm]

Important information for those who will consider buying a Vapcell F60 for assembling large batteries. We assembled the battery in 16s14p format and at a constant current of 50 amperes, the battery overheated 5 times (reached the BMS cutoff of 60 degrees Celsius). The plant subsequently announced that it does not recommend assembling blocks of more than 40 elements from these elements. Quote "we suggest not to make 4000 and 6000mAh capacity for large battery pack , the limited cells for one pack should be less than 20 pcs ; if you want more cells to make pack ,please control 4S if more than 40 pcs , the current should not exceed 15Ah"

This seems really odd. It overheated at 50 amps in 14p configuration? So 3.5 amps per cell? This basically sounds like the company is saying "our cells are unstable, don't use them next to each other"

 

[dominik h]

This will be with all cells. If they are surrounded by other cells , where should the heat go to?
That is the reason why car batteries are cooled with water or have forced air cooling.

 

[harrisonpatm]

This will be with all cells. If they are surrounded by other cells , where should the heat go to?
That is the reason why car batteries are cooled with water or have forced air cooling.

That's fair, but still. 3.5A is only 0.6C discharge for this cell. And it still gets to 60C at 0.6C?

 

[dominik h]

Sounds a bit to low to me, but do we know if the pack can get rid of the generated heat and how the cell connection was made.

 

[harrisonpatm]

Sounds a bit to low to me, but do we know if the pack can get rid of the generated heat and how the cell connection was made.

Those are all important variables, to be sure, but my point is that at 0.6c, a good cell really shouldn't be getting that warm in the first place.

 

[ediskrad]

The assembly is fully open. The ambient temperature is 23 degrees

 

[Pajda]

Thanks for the pics, I wanted to ask, did Vapcell supply you with the original "pink tube" FEB cells as their Vapcell F60?

Btw. I am not suprrised by much because single cell F60 at 1C continuous dischrage finishes around 52°C with initial cell surface and ambient cooling temperature of 25°C in my tests.

So I expect if you made only single charge 0.5C and then 1C discharge without several hours rest time for cell cooldown you definitely ends with 60°C Tcut-off. This has historically been the usual tax for using cells with the highest achievable energy density on the market.

 

[ediskrad]

Thanks for the pics, I wanted to ask, did Vapcell supply you with the original "pink tube" FEB cells as their Vapcell F60?

Btw. I am not suprrised by much because single cell F60 at 1C continuous dischrage finishes around 52°C with initial cell surface and ambient cooling temperature of 25°C in my tests.

So I expect if you made only single charge 0.5C and then 1C discharge without several hours rest time for cell cooldown you definitely ends with 60°C Tcut-off. This has historically been the usual tax for using cells with the highest achievable energy density on the market.

I just directly requested the elements from Vapccell, namely FEB 21700 6000 mAh

 

[ediskrad]

We are currently in the process of selling these items. I have more than 1,500 of them)