C rating for low power 18650s

Hillhater said:
The effect you are seeing is marginal at best, and does not really show on the Lygt test ! (30T )
But none of this explains the Panasonic chart with 6A crossing the 0.5A plot significantly.
I will say again......I have not seen such a discharge effect before.
If you believe it is a “common” effect, please show another chart, from any cell, to verify this. !
The 0.2C tests have always produced the highest Ah capacity results

I am confused by your post. As mentioned, I have seen this effect many times and I uploaded several graphs to Imgur for you to view and offered to email them to you when you said they were low res when you viewed them. My offer stands.

This effect does not require every cell to exhibit the same behavior as the NCR18650A for that effect to exist. My graphs and HKJ’s do not need to be identical for this effect to exist. Whether we’ve seen a particular type of discharge behavior or not has no relevance to whether that behavior exists or not. There are many hundreds of different cells out there and we certainly haven’t seen the performance of them all. I’ve tested over 325 different cells and I still get surprised by new behavior (for me) occasionally.

Check again my Samsung 30T graph. The added capacity at 20A vs 10A easily meets the degree of change seen in the Panasonic cell.

You mentioned 6A vs 0.5A. I don’t claim that the cells I’ve tested showing this effect have higher capacity at a discharge rate of several C versus 0.2C. Only that the effect seen in Pansonic’s graph exists. I don’t know what the high C rate vs. 0.2C capacity numbers look like for the cells I have tested.

We need to ask ourselves why Panasonic would fake their own discharge data? What benefit could there possibly be for this? Especially when it would be incredibly easy to disprove? Isn’t it much more likely that the graph shows the actual discharge behavior for that cell rather than an attempt to fool their customers?

While this effect is real, there is also the possibility of Panasonic’s graph being just a representation of that cell’s behavior and not a result of any particular test. That is, it shows that the capacity at 2C would be higher than expected but we should not take it literally. I haven’t tested that cell so I can’t say how pronounced the effect is but it wouldn’t surprise me that a manufacturer did this. It would helps prospective customers choose which cells they wanted to test in their application, something the manufacturers always recommend doing instead of just looking at the datasheet and discharge graphs.

An easy solution to all this is to get at least a couple of NCR18650A’s and test their capacity at different rates.
 
when i look at data sheets, im looking at
what temperature a cell can deliver 100% capacity
what amps a cell can deliver 100% capacity
theyre not all the same

i like how some of the cells seem to intersect around 3.0v
and by stopping at 2.8v you dont drive the temperatures really high

have you tried "desulfating" any of your old batteries like the hg2,vtc6,30Q that you did the testing on? i think i saw a youtube of a head to head comparison that you did
 
goatman said:
have you tried "desulfating" any of your old batteries like the hg2,vtc6,30Q that you did the testing on? i think i saw a youtube of a head to head comparison that you did

Sorry, was your post for me or the OP?

If for me...no, I have not tried reviving previously tested cells in any way. Desulfating typically involves over-charging and that is a bad thing to do to a li-ion cell. 🙂 Some Li-ion cells that have been in storage for months or years can recover a portion of the capacity they might have lost by cycling the cell normally once or twice at a low to moderate rate.

I have retested plenty of cells though so they do get recharged, but normally. I have had many fail while in storage too. Probably because of the damage caused by my testing. Cells I haven’t abused are typically fine even after years of storage (and the occasional addition of some charge).
 
Camlight,..
I respect you experience and comments on the tests you have posted,..
.....and i am not arguing that those “ overlap” plots you see at high discharges are not real.
Check again my Samsung 30T graph. The added capacity at 20A vs 10A easily meets the degree of change seen in the Panasonic cell.
Unfortunately, the Ltgt test data you also posted for the 30T, does not agree with your own result ?
You mentioned 6A vs 0.5A. I don’t claim that the cells I’ve tested showing this effect have higher capacity at a discharge rate of several C versus 0.2C. Only that the effect seen in Pansonic’s graph exists. I don’t know what the high C rate vs. 0.2C capacity numbers look like for the cells I have tested.
No!..the effect Panasonic shows is higher Ah capacity at 2C than at 0.2C.
None of the other charts posted show that effect.

All i am saying is ........can you, or anyone, show a reputable discharge test on a similar 18650 cell , where the Ah capacity is greater at any discharge, than it is at 0.2C .?
 
Hillhater,
Panasonic 18650A is old cell, old technology. Maybe it is only cell with such pronounced divergence. Moreover, 18650A graph doesn´t look like real measurement. Capacity at 2C isn´t so much higher in newer specification from 2012.
Panasonic 18650A spec 2012 version.jpg

We can see similar tendency at some cells. Compare please Panasonic 18650B (DCIR 50 miliohm), Sanyo GA (DCIR 34 – 35 miliohm), Samsung 30Q (DCIR 18 – 19 miliohm) and LG HG2 (DCIR 20 – 21 miliohm). All DCIR measured according IEC61960-3 at 50 % SOC and 25 ±1 °C. 30Q and HG2 are completely different than Panasonic 18650A and GA.

Panasonic 18650B.jpg
Sanyo GA 1A 2A 3A  4,2-2,5V.jpg
Samsung 30Q.jpg
LG HG2 No2.jpg

Some remarks on Thunderheart and Lygte graphs :
Unfortunately, we cannot rely on THUNDEHEART´s results. His equipment has very probably the same error I had on my ZKETECH, i.e. wrong measurement in 0 – 2 A range resulting in erroneous higher results within lower amp range.
https://endless-sphere.com/forums/viewtopic.php?f=14&t=98081&p=1440617#p1440617
Further he was not using truly 4 wire measurement and has no indication on ambient temperature.
Lygte has no mention about ambient temperature and his tests end at 2,8 V.
 
CamLight said:
goatman said:
have you tried "desulfating" any of your old batteries like the hg2,vtc6,30Q that you did the testing on? i think i saw a youtube of a head to head comparison that you did

Sorry, was your post for me or the OP?

If for me...no, I have not tried reviving previously tested cells in any way. Desulfating typically involves over-charging and that is a bad thing to do to a li-ion cell. 🙂 Some Li-ion cells that have been in storage for months or years can recover a portion of the capacity they might have lost by cycling the cell normally once or twice at a low to moderate rate.

I have retested plenty of cells though so they do get recharged, but normally. I have had many fail while in storage too. Probably because of the damage caused by my testing. Cells I haven’t abused are typically fine even after years of storage (and the occasional addition of some charge).

yes for you, i just say "desulfate" because i dont know what to call it. heres a thread i did 30q,40t,vtc6 but i dont have a hg2 to try it on
i end my cycle tests at 3.2v, you cycle to 2.8v, im curious if it still works
https://endless-sphere.com/forums/viewtopic.php?f=14&t=109601
 
Docware,
Yes , the 18550 A. (And B) are early generation cells such that i have been unable to find any other discharge test results for the 3100mA “A” cell ???
Thanks for posting more discharge curves, but ..
... None have a 0.2C discharge plot
...none are the Pana “A” 3100 cell
...none actually show even the 1.0A plot capacity being exceeded noticably by higher discharge rates
...so that Panasonic “Spec” chart is still unique ?
..and as you say, it does not look like a “real “ dat plot, so i wonder if it is just a marketing “image” ?
 
Hillhater said:
Camlight,..
I respect you experience and comments on the tests you have posted,..
.....and i am not arguing that those “ overlap” plots you see at high discharges are not real.

I misunderstood, thank you for mentioning that.

Hillhater said:
Unfortunately, the Ltgt test data you also posted for the 30T, does not agree with your own result ?

I disagree. They both show higher capacity at a higher discharge level vs a lower discharge level. The results cannot be expected to be identical since we used cells that could have been different grades, manufactured at very different times, and stored under very different conditions.

Hillhater said:
All i am saying is ........can you, or anyone, show a reputable discharge test on a similar 18650 cell , where the Ah capacity is greater at any discharge, than it is at 0.2C .?

Thank you for saying that! I know what you were saying was clear for you but it was not for me.
I don’t plot my capacity test results, they're measured separately, and just wouldn’t have the time to look up the data for a big bunch of cells. I’ve only noticed the effect occurring between two higher discharge current levels though, not between two lower levels.
 
Hillhater said:
..and as you say, it does not look like a “real “ dat plot, so i wonder if it is just a marketing “image” ?

Could be. Created to show that an effect like that exists but not representing the actual average of that cell?
Is that first Panasonic graph (with the pronounced effect) from a technical presentation or marketing document and not the datasheet?

The date of release for the documents can be important too. The graph in question might have been released first as a representation of performance and then a formal datasheet released later with those other graphs showing actual performance?
 
goatman said:
yes for you, i just say "desulfate" because i dont know what to call it. heres a thread i did 30q,40t,vtc6 but i dont have a hg2 to try it on
i end my cycle tests at 3.2v, you cycle to 2.8v, im curious if it still works
https://endless-sphere.com/forums/viewtopic.php?f=14&t=109601
We should continue this elsewhere but you can often partially recover lost capacity (from extended storage, not lost by usage) by cycling a li-ion cell once or twice. I don’t remember the mechanism at work but some reversal of oxidation of the electrolyte rings a bell. You can Google Li-ion aging mechanisms and check out some of the academic and industry research papers.
 
I was curious what is Panasonic 18650B real discharge characteristic.

Panasonic official data list :

Discharge Rate Characteristic for NCR18650B.jpg


My results - I used 5A because of specification 4,9 A max continuous discharging current :

Panasonic 18650B No2  0,2C - 1A - 5A.jpg
Panasonic 18650B No2  0,625A  cell surface warming.jpg
Panasonic 18650B No2  5A  cell surface warming.jpg
 
I haven't done a decent test of the Panasonic NCR18650BD cells I bought, but they check out for capacity at 0.2 C.
DCIR is an average of about 60 mOhms.
 
Docware,..
.. on the 5A discharge temperature trace,.....
..is the 38min peak at the 2.5v cut off ?
....or is that cell still discharging after the 38 min peak temp point..?
Ditto for the 0.62A temp trace at 300 mins ?
Can you correlate the temperature with the equivalent voltage, or time with discharged capacity ?
 
Cell surface temperature appears to correlate with SOC as expected - high SOC and low SOC have higher DCIR values, so the rapid heating is consistent at those parts of the discharge curve.
 
I finally found the Lygt test for the ncr18650, 3100mAh cell..
A little different to the “Official” trace.. ( but at 0.5 and 5.0A discharges )
fCEI2P.jpg
 
Hillhater said:
I finally found the Lygt test for the ncr18650, 3100mAh cell..
A little different to the “Official” trace.. ( but at 0.5 and 5.0A discharges )
Thanks for posting that.
Hard to tell if the traces would cross and, if so, by how much. Seems like they would IMO but can’t assume the same slope for both.
 
Hillhater said:
.. on the 5A discharge temperature trace,.....
..is the 38min peak at the 2.5v cut off ?
....or is that cell still discharging after the 38 min peak temp point..?
Ditto for the 0.62A temp trace at 300 mins ?

In fact 2,5 V cut-off happened between 37 and 38 min. However temperature logging time was set to 1 minute, the graph didn´t catch the temperature peak. Temperature was already going down at 38 min.
For 0,625A peak temp corresponds to 2,5 V cut-off.
 
Hillhater said:
I finally found the Lygt test for the ncr18650, 3100mAh cell..
A little different to the “Official” trace.. ( but at 0.5 and 5.0A discharges )

Not sure how did you achieve to find this test. :)
It is protected cell with protection circuit installed. Moreover the test is from the time, when lygte had non true 4 wire measurement. Those are reasons for the different discharge characteristics, mainly at 5A.
 
Back
Top