SAMSUNG INR18650-29E cycle test.

[Wheazel]

I have 29E cells laying around that are bought in november/december 2014. But maybe that would cause more questions than answers if the cells are not from the same place/time/batch/source. The advantage would be that postage from Sundsvall to Lund wouldnt be more than a day or two.

 

[electricbike]

Ok, back from my holiday now. I charged all 4 cells before i left and now after 1 week i measured the cells again.
Cell-1: 11 mv
Cell-2: 12 mv
Cell-3: 10 mv
Cell-4: 9 mv

It seems that all cell is ok so i will se if there is some problem with the charger that cell 4 was using.
Will start all cycle test soon again.

 

[electricbike]

The test is started again, i hade no time to do it until now.

 

[electricbike]

The first uppdate after aprox 2 week rest of the cell. I have made some minor change, i do 50 cycle insteed of 10 and do a avarage value of the last 10 saved cycle.
It look a litle strange at the cell 4 graph but we will se how the test evolve.

 

[DrkAngel]

Cell 4 test on same or different device?

 

[Hillhater]

I bit late to catch up with this, but one question.
..are you taking the capacity reading (mAhr) from the discharge data , or the charge data ?

 

[electricbike]

Drkangel: yes cell 4 is on the same charger.
Hillhater: The measurment is on discharge.

Test is uppdated 50+ cycles.

 

[electricbike]

Test is uppdated 50+ cycles.

 

[Hillhater]

Well, so far these results seem at least to debunk the theory that charging to 4.2 volts will reduce cell life !

 

[DrkAngel]

Well, so far these results seem at least to debunk the theory that charging to 4.2 volts will reduce cell life !

There is no data-"results" to support that.
Present tests do not provide any comparable data, no commonalities that support comparison! ... ?

Only "results" (I would classify as preliminary speculation) seem to indicate that higher charged voltage is similarly damaging as higher charge rate ... ?

 

[Wheazel]

This thread is helping me make batterydecisions! Had a bunch of 25r and 29e laying around, with this bata I will go for the 29E for this particular application. Thanks for doing this!

 

[electricbike]

Uppdated +150 cycles.

 

[Wheazel]

Intresting, they seem to all do slightly better than the extrapolation.

 

[Wheazel]

Any more words on this?

 

[DrkAngel]

Cycle test of Samsung INR18650-29E.
4 cells under test with different parameters.

Cell Nr 1: discharge 2,7A. charge 1,9A. 4,15 V - 3,10 V

Cell Nr 2: discharge 2,7A. charge 0,8A. 4,20 V - 2,70 V

Cell Nr 3: discharge 2,7A. charge 0,8A. 4,20 V - 3,70 V

Cell Nr 4: discharge 2,7A. charge 0,8A. 4,05 V - 3,55 V

Thread "cleanup" ...

IMO - Test parameters were too differentiated to draw much in the way of life enhancing-diminishing data-conclusions.

Map representation of test cell voltages and usable capacity.

 

[Offroader]

This is why you need to run every 50th cycle or so a full discharge of the cell and graph it, at the same exact discharge rate. This is the only way to compare the different tests.

These tests also need to be done in a temperature controlled environment, because even small temp differences have large effects on capacity.

Without these two important factors the tests won't tell much and too hard to compare.

But I know, I know, I should then do my own tests.. But I am running a test right now with a lot of cells to see how storing cells at different charge rates and temps cause capacity loss after a year.

 

[Wheazel]

I agree that to compare tests there are a lot of more things to consider. Not only mentioned stuff but also what testing equipment is used etc.
This aside, these tests are still very valuable about how this particular cell type behave over time.
With data from trends you can extrapolate the "gaps" and foresee/estimate cell behavior in your own application.
And this is most often enough for good decisions.

 

[Hillhater]

Thread "cleanup" ...

IMO - Test parameters were too differentiated to draw much in the way of life enhancing-diminishing data-conclusions.

IMO..non scientific analysis of results...
1) these cells have nothing worth using below 3.2v
2) there is no apparent penalty for charging to 4.20v..vs 4.15v..but some capacity to be gained.

 

[DrkAngel]

2) there is no apparent penalty for charging to 4.20v..vs 4.15v..but some capacity to be gained.

Cell #1 - Cell Nr 1: discharge 2,7A. charge 1,9A. 4,15 V - 3,10 V
Seems to suffer the same deterioration as
Cell #2 - Cell Nr 2: discharge 2,7A. charge 0,8A. 4,20 V - 2,70 V
While it might appear, to the unobservant, that "there is no apparent penalty for charging to 4.20v..vs 4.15v."
Not considered is the fact that the "charging to 4.15V" cell suffers known deterioration due to higher than recommended charge rate!
(1.9A vs "standard 1.375A")



So, to me, it is obvious that this higher charged voltage is similarly as damaging as this higher charge rate.

#1 - Higher charge rate is known to damage capacity and cycle life.
#2 - Higher charged voltage is known to damage capacity and cycle life.

 

[redilast]

Thread "cleanup" ...

IMO - Test parameters were too differentiated to draw much in the way of life enhancing-diminishing data-conclusions.

IMO..non scientific analysis of results...
1) these cells have nothing worth using below 3.2v
2) there is no apparent penalty for charging to 4.20v..vs 4.15v..but some capacity to be gained.

1) Hit them with 10A and there is plenty worth using below 3.2v.

2) The 4.2 vs 4.15v tests are incomparable. One uses a significantly higher charge rate, higher charge rates above 4v are very bad for most Li-ions.

 

[Hillhater]

..fair comments guys, but..
...There is no data to quantify any capacity loss due to the higher charge rate.
..and true , Redilast,...those capacity maps are only representative for the one discharge rate ...0.2 A
..and the one used (0.2A discharge) , is not relavent to these tests.
DA,..do yo have any higher rate discharge capacity maps ?

 

[redilast]

..fair comments guys, but..
...There is no data to quantify any capacity loss due to the higher charge rate.
..and true , Redilast,...those capacity maps are only representative for the one discharge rate ...0.2 A
..and the one used (0.2A discharge) , is not relavent to these tests.
DA,..do yo have any higher rate discharge capacity maps ?

There is another thread where someone did some cycle life testing on some cells with like a 1C charge rate, and there was considerable capacity loss after even just 100 cycles. Like 50% capacity loss. From everything I've read and researched in the past, what is most damaging is right before the current starts to taper off (e.g. on the high voltage end of the CC phase before it switches to CV). 1C charging would probably be fine for cycle life if it wasn't using CC/CV, but instead decreased current above 4v or so. For example CC to 4v, then 75% reduction in charge rate to 4.2v then CV with gradual taper down of current till a low termination current in the 50-100mA range. Maybe someone recalls the thread, or heck it could even be somewhere in this one on another page...

As for higher discharge tests, you can see 7A tests here of the 29E http://lygte-info.dk/review/batteries2012/Common18650comparator.php

Select The Samsung 29E from the drop-down then select 7A. Around 20-25% capacity remaining at 3.1v under a 7A load for example.

 

[DrkAngel]

..fair comments guys, but..
...There is no data to quantify any capacity loss due to the higher charge rate.
..and true , Redilast,...those capacity maps are only representative for the one discharge rate ...0.2 A
..and the one used (0.2A discharge) , is not relavent to these tests.
DA,..do yo have any higher rate discharge capacity maps ?

Capacity loss from higher charge voltage, mirrors capacity loss from known damaging higher charge rate! = data indicates capacity loss due to the higher charge rate.
My "Capacity maps" are tasked to indicate capacity at static voltages and so are created at the minimal discharge rates to reflect best accuracy.
Capacity maps are produced to help determine optimal charge voltages
and
to help determine optimal discharged voltage ... not discharge voltage, discharged voltage.
Resultant voltage after discharge concluded.
Discharge voltage would be continuously variable, unquantifiable except for at a specific motor x rpm + controller x throttle position + battery condition x actual capacity

 

[Offroader]

..fair comments guys, but..
...There is no data to quantify any capacity loss due to the higher charge rate.
..and true , Redilast,...those capacity maps are only representative for the one discharge rate ...0.2 A
..and the one used (0.2A discharge) , is not relavent to these tests.
DA,..do yo have any higher rate discharge capacity maps ?

There is another thread where someone did some cycle life testing on some cells with like a 1C charge rate, and there was considerable capacity loss after even just 100 cycles. Like 50% capacity loss. From everything I've read and researched in the past, what is most damaging is right before the current starts to taper off (e.g. on the high voltage end of the CC phase before it switches to CV). 1C charging would probably be fine for cycle life if it wasn't using CC/CV, but instead decreased current above 4v or so. For example CC to 4v, then 75% reduction in charge rate to 4.2v then CV with gradual taper down of current till a low termination current in the 50-100mA range. Maybe someone recalls the thread, or heck it could even be somewhere in this one on another page...

As for higher discharge tests, you can see 7A tests here of the 29E http://lygte-info.dk/review/batteries2012/Common18650comparator.php


Select The Samsung 29E from the drop-down then select 7A. Around 20-25% capacity remaining at 3.1v under a 7A load for example.

I'd like to see this thread about charging but I believe it is understood that higher charging rates, even if below the battery spec, seems to do a lot of damage.

This is why I usually charge my cells at around .2C charge. I don't care most of the time if it takes 2 hours of 8 hours to charge my pack.

My 18650GA samsung cells are rated at a 1.6 amp charge rate per cell (about .5 C charge), but since I have 14 of them in parallel, I don't care about charging them at .7 amps each cells. Still gives me a 700 watt charge rate because my pack is so large. It also still heats the pack up by a decent amount at that low charge rate.

But if one was to use these 18650 GA 3500 cells and charge them at 1.5 amp per cell, I agree you would get diminished life from the. I only know of one person who charged them above spec at like 3 amps per cell, and he got greatly diminished capacity.

Discharge at higher rates is probably not that much better, but most of us pulse the discharge when riding. If one was to run it full rated discharge, even half the rated discharge, you would probably get reduced capacity quickly, and a very hot battery.

 

[Hillhater]

.....
2) The 4.2 vs 4.15v tests are incomparable. One uses a significantly higher charge rate, higher charge rates above 4v are very bad for most Li-ions.

....Capacity loss from higher charge voltage, mirrors capacity loss from known damaging higher charge rate!....

but the "cell 3" result does not indicate any reduced capacity with the 4.2v charge level ...( but with the shallow discharge its a unreliable conclusion )
What would be useful now , is to do as Offroader suggested and perform a full capacity test ( 4.2 -2.5v ) at 1C , to compare the effects of these cycle trials better .