Discharge after capacity testing?

I am still in the beginning throws of testing capacity of ~650 cells and growing daily with a single Opus C3100 charger, and working on an extreme budget. After testing the cells and letting them rest for about a week I have been discharging them to ~3.40 - 3.5 volts and placing them in the bin according to capacity. So my question is, it is worth the time and effort to discharge them? It will be awhile before I am ready to start assembling the packs (either 24v or 36v at this time, long term plan is to replace the sealed lead acid batteries in either an APC or Tripp-Lite with a BMS).

After charging them, are you letting them sit at full charge for a few weeks to weed out most of the self dischargers?

Discharging to 3.5V (which 3.7V would be sufficient) and letting them sit is fine for long term storage, like 6 months to a year or longer.

Not yet, mostly looking at capacity at this time, however I have been keeping track of the cells, noting the voltage as it was salvaged from the pack, the voltage before testing, and the resting voltage after the capacity test with the date. Maybe a little overkill on what all I am keeping track of on a per-cell basis, but it may prove useful in the future. It will be a good 6 months before I will have any free funding to start building the packs.

Nope, not overkill. Good records. However, you need to do the resting voltage with full charge. If you drop the voltage down to 3.6V, you will have a hard time finding those SD cells. Might as well get them out of the way now.

And actually, the order should be:

1) Full charge cell
2) Sit for 2 - 3 weeks to check for SD
3) Drops more than .02V/day, considered SD cell and put in separate bins
* If they stabilize around 4.0V or even 3.9V, they could be used in small applications, like flashlights and such
4) Remaining cells get recharged to full, then discharge capacity tested. If they are within spec, keep. If they are quite short, toss
5) Recharge to full charge and let cells sit for another 2 - 3 weeks, repeating step 3
6) Cells that pass step 5 then can be discharged to storage voltage (about 3.7V) if going to store for many months to a year or longer. Store in cool dry place if possible. At least in a sealed container if humidity is an issue.

This is how I would do the steps. There may be some other steps that could be added as well to get even better results on good cells

If you focus on capacity test "first", then you could potentially be wasting a lot of time, energy and effort on cells that are SD's. Why waste that on the ones you are going to toss anyways?

I know I should do the SD charge, but like I said I got a lot of time at the moment while planning on the packs. Having a ton of kids certainly does not help on the funds at the moment, and the next one will be here in over a month from now. I would like to have as much data at the moment on the cells to possibly assist with issues with the packs down the road. It's a learning experience for sure.

I would like to follow up with a IR test as well in the near future and test those that have already been put in the spreadsheet, will need to do some searching for the resistances (maybe suggest putting in the proper resistances in the cell database for ease). Looking at the YR1030 from vapecell, reasonably priced for what I need.

Plus one on Korisan.
Charge them up to 4.20, wait 4 weeks(very important), then a discharge test....
After 4 weeks:
The time saving bonus after the four weeks, if there is a cell that dropped below 3.9 or lower is a sd'dr.
Trow them out
So you don't have to take them into further testing.
With the discharge test after the four weeks, you will weed out every bad cell.
Everything that is below 80% from there original capacity, is not worth the effort anymore!

I would also recommend to buy a ir tester, the yr1030 is the best for this, and dive into the world of ir.
It saves more time when you test your cells first on ir, then charge, wiat 4 weeks then discharge.
Then you will have the best cells for your pw and NO dubble work!

Voltage for long term storage: 3.3-3.4v, cous on this point that there is nothing that can harm the cell anymore.
For short term storage it is best to have them at 3.6v at this point the cells chemistry is at there best balance.

First learn to walk, then learn to run, meaning: forget your 6 months goal, spend more time in the beginning with your cells, to save a lot of time in the end, and create better packs.

Search on this forum for Wolf, The Wolfmandid not reinvent the wheel he invented it.

My testing way(remember that there are more ways to rome!) after getting my cells.
1 with the ir meter, test for v and ir, odd ir and below 2 volts are out.
2 charge to 4.2 volt.
3 wait 4 weeks.
4 check for v, every cell below 3.9 is out
5 discharge to 2.8v and calculate the remaining capacity,below 80% soh is out.
6 charge them up to 3.35v for storage.
7 sort them into 100mah trays to build your packs.
8 solder
9 enjoy
10 sit back and relax

For my first 1000-2000 Cells i have pre tested them for voltage drop. The ones who dropped (3,9-4,05) got marked with a "V". I tested them und let them sit for at least a Month, but half the cells have no more voltage sagging after that. There are even ones who have 4,15V after 8-9 months. So i dont pre test them any more. Charge, discharge, store (1-2 months) voltage test.

@nickydw.
You will want to know the SOH in %
Especially for the "german punktlighet" (sorry if it was grundligheid).
You can get away with that, but if you have one cell in a pack of 100 you will not be able to tell what is going on.
So it is a waste of work.
I have a "secret test setup" that only contains cells that are not dropping below 4.00.
Not looking good, especially in the balancing akt and ir, there are some sd also.
test cells: icr18650 26h, 25 cells a pack x 14...
Again you can get away with that for the next two years....just saying

what is the difference between checking the voltage loss before or after the capacity measurement? of course the cell will be fully charged after test.

Checking the voltage right after charge gives you your starting point. If you pull the cell off and just assume it's 4.2V, you could be wrong. Also, some cells start SD immediately after coming off the charger. I've seen some drop almost 0.1V within minutes of pulling off the charger. That cell gets binned and not even further tested, at least for powerwall applications.
Also, some chargers will stop at 4.18, 4.21, 4.2, 4.22. These are all safe voltages and can vary from charger to charger and from slot to slot. So a baseline voltage reading is needed to know how much the cell voltage dropped after resting.

On average, a cell that drops 0.01V/day is a good cell. 0.05V/day is too much. And you don't want to loose more than about 0.1V - 0.15V overall after a month of resting. If it's more than that, it's on its way to dieing

Also to add: if a cell take to long to charge it could also be a potential sd.
If you charge a 2000mah cell with 1amp a hour and it takes 4 hour to complete then there is something wrong with that cell.
Inlc also heating up.
Those cells must be discharged immediately to 0 volt and in the bin for the recycler.

Now most diy chargers like the tc40xx can charge 1ah but they must have the proper power supply for it also.
So if the max is 500mah that the charger is getting, then a 2000mah cell must have 5 hours to complete full state.
Make sure you got power enough to spare, when charging cells

Before you consider a sd after the cell comes from the charger, give the chemistry some rest so it can settle, usually it takes one hour.
To make sure just wait 4 weeks.
To make it more complicated on the charging part: its not only from charger to charger or from slot to slot.
Some cells simply dont want to be charged to 4.15v and the other will take charges up to 4.25 without pain.

Arnt we having fun? :huh: :s

Even though I didn't do it this way, I would recommend charging and testing for voltage drop FIRST before capacity testing for 2 reasons: (1) It will help you weed out self draining cells and save you the time you would spend testing those cells; and (2) keeping cells at full charge for extended periods of time WILL damage your cells. So why test twice? I made that mistake and let cells sit at full charge for between 8 months and 1 year before testing AGAIN and found out I lost like 3% of my good cells and a whole bunch of time retesting. So after you test for self-dischargers, do a capacity test, and then bring those cells to a uniform storage voltage (I would do 3.4V but most would do 3.7V). Keeping the voltage low and uniform will aid in pack building. You don't want to solder/spot weld a fully charged cell in case you make a mistake (it happens).

But doing that way includes more moments to work with it manually.

For me doing it with Opus its just one moment and then 2 weeks later i know if self discharge is ok.,

Ie
1. Put into opus let it run capacity test and also tell me if IR is ok (heat)
2. Let it sit and then meassure for self discharge

The other way would be
1. Charge it up
2. Let it sit and then test for self discharge
3. Do capacity test in opus or even worst charge them up adn then put them into another tester...

For me the manual labour is more costly than anything else

interesting i have built a 11kw powerwall.i have been reading the messageboard and maybe i have made a mistake on cell selection.The method you use will result in a very high quality battery bank,but with a lot of waistage.What i did was set up a bank of 14 TP4054 chargers and used 5v computer supply charged cells untill modual said full charge(this is automatic as light changes colour).Any cells that is warm is discarded.This is were i differ i leave for 24 hours any cell under 4v is discarded as it is a SD CELL.I then immedily discharge test to 2.8v and write on cell capacity.For discharge i use these chines discharge moduals that are resonably accurate for my needs.i have some weak banks on my powerwall so add some batteries to balance out.It seems your way is a good one to increase the quality of power bank with probably some small cost but i think might be way forward to increase life of the battery bank overall.Any comments will be read as i want to increase bank life .

To increase battery bank life = get more cells and dont stress them as much

For having a battery bank that will last from start make sure to weed out ALL cells questionable. If they cant sustain sitting fully charged for several months they will for sure not last in a powerwall being abused

Thats how easy it is. If they dont pass first test they are out.

@Korishan that's my usual way to do it, except 6) . I don't discharge them.

For my process:
I am using tp4056 to charge and zb206 to capacity test.
The tp4056 are much cheaper (from memory around AU$0.45) than the zb206 (about AU$11.50)
I will have more tp4056 and do the charging/self discharge test first and only capacity test the cells which pass the self discharge test.
This saves the more expensive capacity testers from being used on cells which may fail SD testing.

Generic said:

Even though I didn't do it this way, I would recommend charging and testing for voltage drop FIRST before capacity testing for 2 reasons: (1) It will help you weed out self draining cells and save you the time you would spend testing those cells; and (2) keeping cells at full charge for extended periods of time WILL damage your cells. So why test twice? I made that mistake and let cells sit at full charge for between 8 months and 1 year before testing AGAIN and found out I lost like 3% of my good cells and a whole bunch of time retesting. So after you test for self-dischargers, do a capacity test, and then bring those cells to a uniform storage voltage (I would do 3.4V but most would do 3.7V). Keeping the voltage low and uniform will aid in pack building. You don't want to solder/spot weld a fully charged cell in case you make a mistake (it happens).

Click to expand...

I'm in a similar position, I'm pretty new to this I salvaged600 x 18650 cells from old laptop batteries,fullycharged removing heaters and leaving the rest for 2-3 weeks to weed out the any that SD then capacity tested using the NOR test on my LiitoKala Lii-500's

Now I've realised I wont get to build all of these in to packs for a while and I have 600 cells at 4.2v Why doesn't the NOR test charge to a storage voltage rather than a full charge surely that wouldbe more useful and quicker?

Any advice on how best to get this many cells down to 3.7v to store for a while

I'm surprised at the amount of people that leave their cells charged at full voltage for month(s). There is quite a body of work out there that says don't do it as you are needlessly losing capacity. Besides the many articles at BatteryUniversity.com (including this and this)there is this white paper that has done quite a bit of empirical work -

https://www.researchgate.net/public..._Battery_Degradation_for_Cell_Life_Assessment

Diggsut... Yeah you can loose up to couple of % per year doing it BUT.... Most bigger solar installs you will also have the powerwall fully charged for up to 8 hours a day.

So if we look at the numbers. Lets say you have used cells and you let them sit at 15c that I do... Lets say I leave them for 3 months. If you look at that white paper you will see that I loose a whopping 1% or so.

So if we take this further and if you read that white paper you send you will see that sitting at lower SoC does not save you that much in terms of degradation during those 3 months. If you add how much you actually degrade using the cells later on especially for those pushing cells harder you degrade ALOT faster doing that

If you pull all those numbers on the table this is the result:
* Sitting at full charge some month does not kill the cells
* Spending that extra time discharging and wasting that capacity in the cells cost MORE in time and energy wasted than actually just buying some extra cells to compensate the minimal loss
* Sitting att full charge during that time you can even better weed out self dischargers. That self discharging cells actually saves you more energy than what you have lost letting them sit there comparing using the batteries over some years.
* If a cell actually wont be able to sit at full charge for lets say 3 months... Its not even worth having in a powerwall. Because 3 months full there equals 9 summer months sitting full in my powerwall... And this I cannot do anything about unless i want the battery bank to sit half charged and loose valuable DoD capacity.

Do I have numbers of this? Yes i calculated and tested this. Im not going to write all numbers here because thats a white paper it self

Yes - Leaving them fully charge degrades the batteries slightly faster than middle/close to empty.
BUT just lowering the temperature where they sits do ALOT and really... If thats your biggest concern loosing energy you should consider buiing better wiring or inverter/charger. Thats where you save alot more energy.

Would letting them sitting fully charge cause a "fire" -> NO.

We have a couple of side-tracks here becaue LiPo on other hand are more sensitive and thats because of the case. If LiPo would sit in hard-case its not a problem there either. I have done my fair share of RC last 10+ years and I have many cells sitting for all winter fully charged for testing. Some brands doesnt like it as much when using soft cases since they do wanna puff a bit but so far not problem.
I have killed more cells due to abusing them when in use than leaving them as is


Lastly: If you dont know when you will use them or its a very long time like half a year or years then of course discharge them down to 40-50%. its just better. But for a month or 3 during the build its ok