Results of Capacity Testing 18650s for 1,600+ Cycles

Oz18650

Member
Joined
Oct 2, 2017
Messages
200
Congratulations to Generic for the testing and sharing the results.
I know that this takes a lot of commitment to do manually.
I am only at 140 tests between 2 cells, and while each test is fairly quick, it still adds up.


rev0 said:
Wow, you did all these cycles manually? If you're interested in continuing this sort of testing I could send you an 18650 cycle tester I've been working on (would be an older version of which I have several), details are here: http://rev0.net/index.php?title=CCR It sure seems easier than manually restarting a cycle on these cells and taking down the data.

That is a lovely offer.
It is nice to see someone else supporting generics testing and contributions to the community. I offered to provide an opus tester earlier on, so his device is not tied up.

I have had a quick look at your project and this looks very interesting to me as I also would like to do some repeated testing of cells.

I have started already with a single zb206 and tp4056 and have done 140 tests across 2 cells, so only 70 cycles for each cell.
I do have a bunch of questions that would be answered by being able to automate this testing. From the manual testing I have already done, I have determined that it would really not be feasible for me to do without automation.

I have many cells that are exactly the same type, and the packs have date codes.

I would like to observe how cells behave over many cycles where
-cells are different ages based on manufacture date
-cells that have had very little use
-cells that have had lots of use
-cells have been at zero volts and recovered
-1A charge and discharge
-0.5A charge and discharge
-reduced voltage range

I believe low use vs lots of use could be reasonably estimated based on the state of the packs and devices my cells come from.

I think the 1A vs 0.5A comparison would be a nice contribution to the second life storage community.
In theory the reduced currents and voltage range extend cell life, but being able to show the size of the effect in real life would be great.
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
Oz -Well, the newest Samsungs that Thunderheart has been reviewing quote cycle life to 70%, so I think the actual percentage is a bit arbitrary. After 70 cycles on each cell, are you seeing any degradation?

rev0 - Wow, you've done some very interesting research! Another member mentioned about the ReVolt cells only getting 340 cycles before nosediving, and I couldn't find the results until I clicked your link! I may be interested in your offer, I'll send you a PM.
 

Oz18650

Member
Joined
Oct 2, 2017
Messages
200
Generic said:
After 70 cycles on each cell, are you seeing any degradation?

I don't think I have seen any degradation yet, certainly not that I can spot without analysis of the results.
I just got the lowest result yet, but also the highest result yet a couple of days ago.
 

YAILIYA

New member
Joined
Sep 12, 2019
Messages
2
Overmind said:
I agree with Dallski here, these home testers are good enough for general use but not good enough for professional use.

If I cannot set the exact discharge max current and the minimum voltage specified by the manufacturer, the test is not professionally relevant. Also, none of the testers I have encountered support high discharge currents. That makes most of the tests of high power cells irrelevant. They will test fine at 1A discharge and instant die at 10 or 20A.
I'm going to use the Charge Test function at 1000mAh and record the capacity for each of the cells after each cycle. These will all be full cycle 4.2V to 2.8V discharges, so it's a lot more abuse than a typical powerwall will give these cells, but it will also help speed up the results. I'm going to aim for at least 100 full cycles.
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
Update time! 705 cycles. Chart is self-explanatory. We are now past the 70 or 80% that every datasheet I have ever seen usually stops at.I am also re-uploading the degradation numbers at each 100 cycle interval. I was doing the math with the Wolf method, but was getting over 100% degradation, so I'm reverting to how we all do state of health: tested capacity over initial capacity. I'm starting to wonder if the THLD cell will make it to 805 cycles? I'm also considering changing the Title of this thread to "Long Term Capacity Testing Cells." I think a lot of people who would benefit from this information are skipping over this thread because they have an uninformed negative bias toward generic cells (I used to be one of those people).


image_ertmsf.jpg



image_wjmant.jpg



image_qhyzyf.jpg
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
image_fcqslg.jpg


First cell to be completely degraded: CJ! It has been testing under 100 mAh since Cycle 776, butit is still putting out about 64 mAh for 30 straight cycles. I think I'm done testing this cell at 1000mA, so now that I have an official number for Cycles 801-805 for my data, I'm going to run the cell at 500mAdischarge for Cycle 806 to see what capacity it has at a lower discharge rate. And maybe 300mA for Cycle 807 and 200mA for Cycle 808 if it doesn't take forever. Then after Cycle 808, I'm going to fully charge it at a200mA charge rate, and test its voltage at weekly intervals to see if it self discharges. The THLD cell looks like it's going to be fully degraded pretty soon, too, so I think I will do the same thing to it after 5 or 10 sub-100mAh cycles.

image_cczkmu.jpg

image_qfzntl.jpg


You may notice that the ASO and LG cells seem to be dipping quicker than before, but I don't think they are degrading any faster. I think they are showing lower capacities because the CJ and THLD cells are not generating heat next to them like they did before. I wish I had temperature correction like rev0's equipment, but I don't, and this is one of the downsides of the Opus.

Also, like any true research, data creates more questions than it answers. For instance, the THLD cell started dropping capacity much earlier than the CJ cell, but the rate of decline was flatter. The CJ dropped like a rock, from about 1000mAh to <100mAh in just 75 cycles. Could it be the chemistry? Electrolyte? Cobalt content?Age?


image_dcanbg.jpg


I also wanted to post this chart. It shows how much energy each cell has stored in its lifetime. The CJ cell was able to store 5 kilowatts of energy in its lifetime!

Lastly, I just want to mention that I have received the cell cycler from rev0 and am excited to start another long-term test after I do some other testing like high-drain generics and possibly some end-of-life (50% of original capacity)genuine cells to see their degradation much more quickly than with new cells. We'll see. Stay tuned?
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
Congratulations! You finally wore a cell out.

Glad you are going to be able to have another round. Especially with some level of automation.
 

Oz18650

Member
Joined
Oct 2, 2017
Messages
200
Great work!
I wonder how much the current being drawn compared to the capacity of the cell contributes to the accelerated drop in capacity?
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
Great question, Oz. I do believe if I did this test at 500mA instead of 1000mA, I would have gotten more cycles. Most likely due to a slower chemical reaction within the cell and reduced heat from the slower reaction (and less heat from the Opus). But I can't say for sure because I don't have any proof.

What I can say, though is that I did test the CJ cell at 500mA, 300mA, and 200mA, like I said I would:
66 mAh @1,000mA
330 mAh @500mA
505 mAh @ 300mA
679 mAh @ 200mA

So, I got 10x the capacity by reducing the discharge rate 5x on a "dead" cell. Interesting. So if I had a 50kW powerwall, and ran a coffee maker, toaster oven, microwave, and a window AC all at the same time for a 5kW load, the cell would be doing 200mA and get 679mAh useable. I'd have to be drawing 25kW from the powerwall to only get 66mAh from the "dead" cell.
 

Oz18650

Member
Joined
Oct 2, 2017
Messages
200
Those final test results are very interesting.
I think they might be showing that a cell which no one would ever consider using after a 66mah test result, would actually contribute in a powerwall?!
If that is true, not only would it be contributing, but it would be taking some load off other cells, making them last longer too.
What does everyone else think about that?
 

Wolf

Well-known member
Joined
Sep 25, 2018
Messages
1,278
Oz18650 said:
Those final test results are very interesting.
I think they might be showing that a cell which no one would ever consider using after a 66mah test result, would actually contribute in a powerwall?!
If that is true, not only would it be contributing, but it would be taking some load off other cells, making them last longer too.
What does everyone else think about that?

@Oz18650,


Ozzy I love you right to death but have you seen my charts and tests on how cells behave in a parallel discharge and charge environment?
https://secondlifestorage.com/showthread.php?tid=7846&pid=53209#pid53209
Also the thread on high and low drain cells in parallel.
https://secondlifestorage.com/showthread.php?tid=7873&pid=53600#pid53600
If you haven't OK I forgive you but if you have you must have been goofing off in the class. :p
A low producing cell will actually put more stress on a good cell as the good cell has to take up the slack of the underperformer.

The ideal battery pack would be all cells of the same IR and the same capacity AND the same manufacturer.
But in our reclamation world we have different manufactures different chemistries etc.
So we need to mix and match, but we need to mix and match within certain parameters to have a successful final battery.

The one thing missing from all these cycles is a regular IR check at say every 100 cycles. Nevertheless I already know the outcome. The low mAhcells will have a higher than normal IR. More than likely >90m?.

Ozzy Ozzy Ozzy :huh: :huh: :huh:

Wolf
 

Oz18650

Member
Joined
Oct 2, 2017
Messages
200
:)
Thanks wolf.
Not quite goofing off, but it's hard to take in those threads from my phone.
I will have to study more.
 

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
659
@ oz18650,
We all have to study more, thanks to your work and other we can study and learn more.

Oz18650 said:
Those final test results are very interesting.
I think they might be showing that a cell which no one would ever consider using after a 66mah test result, would actually contribute in a powerwall?!
If that is true, not only would it be contributing, but it would be taking some load off other cells, making them last longer too.
What does everyone else think about that?

Everyone can afford to lose a dime, if everyone on this forum will transfer 10 cents to my account i would have a great christmas vacation.
It is the small bits.

But like wolf said:
"A low producing cell will actually put more stress on a good cell as the good cell has to take up the slack of the underperformer".

But i think that mostly depends on how much you are asking of those cells, 5a a cell or 50ma a cell(below 66ma)
That's why i regret to bring the low amp cells to the recycler aldo they where 90%soh.

Would like to hear more thoughts

Best
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
Oz, after the 200mA discharge test, it took 8 hours and 34 minutes to charge the cell back up to full. I wish I would have discharged the cell at 200mA and then manually charged the cell on the Opus to count how many mAh went into the cell. I'm guessing it was pretty inefficient due to high IR. Just goes to show you how important it is to test cells at 1000mA and not at 500mA. And also, even if it could give you 679mAh at a 200mA discharge rate, it is end of life, anyway, so it would start putting more stress on the good cells, as Wolf has mentioned.

Bubba, now that the CJ cell is done, I'm going to check it's voltage at weekly intervals to see if it self-discharges. My guess is that it will, but not very quickly. Not all low state of health cells self-discharge. I also don't want to interrupt this test to check for self-discharge, because I guarantee you that my motivation for doing this test will be gone if I stop doing it for a week or more. I did have a couple of crazy work days where I went 48 hours without testing these cells at the end of October, and I didn't see any self-discharge at that time, but that was only 2 days rest for the cells. Lastly, the CJ cell was fully charged about 30 hours ago and is reading 4.16V. I'll test it in a week to see where it is at then.

Professor Wolf, right on, as usual!! I was going to purchase a legit IR tester like you have, but when I made the switch from used laptop packs to new-old-stock generics, I failed to see the need for it. That being said, I have been watching the initial voltage drop between the charge/discharge cycles, and the ASO and LG cells have been dipping faster and faster, so it's just a matter of time before the dip starts affecting the discharge curve and in turn, the cells' capacity. I'm sure the IR has been growing for at least 300 cycles, maybe more.

100kwh-hunter, I see value in everything, so I have a hard time getting rid of low State-of-Health cells, unlike the self-dischargers or heaters that I have no problem eliminating. But I save low State of Health cells for side projects. You don't want to build a massive powerwall and have a few bad cells ruin it.
 

rev0

Active member
Joined
Oct 3, 2017
Messages
369
I would suggest if you do want to use low capacity/quality cells in a powerwall build, group them together in their own small pack (say 20p) and hook it to the main pack (e.g. 80p+) through its own 20p-pack fuse. Bonus points if you monitor the current into/out of the small pack and the main pack so you can keep track of whether the small pack is falling behind or self discharging (taking more mAH than it can give).
 

Generic

Member
Joined
Oct 23, 2018
Messages
210
Great idea rev0! I was thinking another use-case was to use those cells as a service pack, so if you have to take a pack down for maintenance, you could put a low capacity pack in during the maintenance period so you don't have any real downtime.
 

100kwh-hunter

Active member
Joined
Mar 2, 2019
Messages
659
Generic,
What if you would made a discharge setup with a kwh counter, it don't have to be anything fancy, just a old analog kwh counter will do.
Keep records of that also?
But a cell that is mfg rated at 1600 and is still good with 1525mah after 10 years...yes with the knowledge i have know, i would make them into packs of 20....i think i brought 20kwh+ to the recycler, especially after reading your posts, just saying


Thanks and keep up the good work
 

Wolf

Well-known member
Joined
Sep 25, 2018
Messages
1,278
rev0 said:
I would suggest if you do want to use low capacity/quality cells in a powerwall build, group them together in their own small pack (say 20p) and hook it to the main pack (e.g. 80p+) through its own 20p-pack fuse. Bonus points if you monitor the current into/out of the small pack and the main pack so you can keep track of whether the small pack is falling behind or self discharging (taking more mAH than it can give).
@rev0

The idea of adding a "small" 20p pack oflow capacity/quality cellsto an existing power wall build is somewhat of a problem.
Please see my posts and discussion on how parallel cells act and thehigh and low drain cells mixed.
Think of it this way, You have 1 cell that is a 3000mAh cell with a SOH 95%and you have another cell that is 500mAh at a SOH of 25%you install these cells in parallel and then you ask for them to provide power. Let's say 1A draw for the parallel pair.Mathematically and electricitywe know the 2 cells should provide equal power of 500mA if theyare matched. We also know that the3000mAh cell will provide 500mA for 6 hrs. and the 500mAh cell will provide 500mA for an hr. What does that mean? Well I can tell you from experience that the 3000mAh cell will provide most of the 1A draw and the 500mAhcell due to its higher IR and lowSOH will just give up as little as it can and cause the 3000mAh cell to do most of the work.
After the discharge and when equalization kicks in the 3000mAh cell will recover very quickly but unfortunately give up a lot of its recoverable energy to charge the now depleted 500mAh cell.
As the 2 cells are bound by Voltage not mA the Voltage will always have to be the same no matter what the draw is.
Think of itlike putting a 1.5V D cell and a 1.5 AAA cell in parallel. Which one will power the flashlight more and when you switch it off what will happen to the AAA cell.

Now take that scenario to the next level. You have an 80p pack with relatively good SOHcells 80% lets say and all cellswith reasonable IR.
Knowing from the study I have done on parallel cell interaction we know that there will be some cell interaction to maintain an even voltage for the pack and there will be a minimal current flow between all cells as the pack balances internally. That can take sometimes 5 to 8hrs. If not longer.Now once that 80p pack is internally balanced you in essence have 1 cell at a nominal~3.8 V. Nowif the 80p pack has80 cell withan average of 2200mAh that battery becomes a "single" cell of 176 Ah. Now we introduce a parallel cell pack of 20p at 500 mAh per cell.OK its an extra 10Ah but what does that do for us?
All cells in parallel act as 1 battery cell. We now need to go through the initial "pack" balance again and there is the potential of a large influx of current going to the 20p pack. rev0 mentioned a fuse separating the 2 which definitely would be a good idea as the potential of a large current exchange is definitely possible. If both packs are the same voltage the initial balancing current between the 80p and 20p packwould be reasonable and probably not cause a problem.
The problem comes in during the discharge. The 80p pack will more that happily give up its mAhsas it is being discharged as will the 20p pack. As the voltage drops both packs will oscillate between each other one taking from the other as is necessary. At the DOD cutoff lets say 3.2V the 80p pack will recover almost instantaneously as theSOH of the batteries is good and force the 20p pack to come to compliance causing a large influx of current into the 20p pack as to maintain the balance voltage. The resulting end effect is that the 80p pack will continue to discharge to bring the 20p pack into electrical equilibrium and compliance. End result is large currents are exchanged between the 2 packs and in my opinion (based on the studies I have performed) not a good idea.

I think you would find that byadding a substandard pack to a good working pack would do more harm than good and actually cause a degradationof the system and no appreciable performance gains would be realized.
Wolf
 
Top