Recovering low voltage cells

What do you use to recover them?

I liked the idea of @krlenjuska to raise them to 3.7v with low amperage and let them sit. But what do you use, that has that variable current and can set the target voltage?

Resistor via another cell to above 2v and then into the Opus that do the rest.


Nothing special.

Just Keep the current low.

if single cells/cells in parallel
imax 1s mode
icharger X6 1s mode
icharger x8 1s mode
if cells in series
up to 6s imax, icharger X6
up to 8s Icharger X8 I recovered 6 24v 7s batteries with my x8 used .02Ato charge to 21v then 2A to24v then 4A to 28.5v
later floyd

SeRiusRod said:

What do you use to recover them?

I liked the idea of @krlenjuska to raise them to 3.7v with low amperage and let them sit. But what do you use, that has that variable current and can set the target voltage?

Click to expand...

I believe TP5100 will pre-charge and adjust in steps as the cell returns to the normal range.

A while back, I processed several hundred'raw' Power2Spare medical packs - with 8 x NCR18650A(s) in them. Many were 0v/lowv - and I used @DIY Tech & Repairs resistor approach (youtube above - thanks to @Daromer for this). I was able to recover a majority of them to be able tocharge up in my OPUSs- however - many of them got HOT (e.g. too hot to touch) and had to toss them.

Not sure if it's a 0v thing or a NCR18650A thing - but I'd advise paying particular attention to heaters (if they start as 0v/lowv), particularly asas they near full charge.

OffGridInTheCity said:

A while back, I processed several hundred[...] medical packs. Many were 0v/lowv - and [...] I was able to recover a majority of them to be able tocharge up in my OPUSs- however - many of them got HOT (e.g. too hot to touch) and had to toss them.

Click to expand...

Safety warning: generally manufacturers warn to not charge extremely overdischarged Li-ion cells, i.e. cells <1.5V, eg. 0V, (though some use a lower 1.0V threshold). Charging can lead to serious safety problems (briefly: copper can dissolve at low voltages and then during charge replate into dendrites, possibly leading to dangerous internal shorts at some future time - often without any warning). For some very old cells (lacking modern safety improvements) it may not even be safe to charge them if below 2.0V.

While there have been safety improvements over the years to reduce such risks (e.g. browsing patents reveals chemistry improvements to suppress copper dissolution, and ceramic separators to suppress or localize internal shorts) there is generally no way for consumers to know which improvements apply to particular cell models, since such information is not typically provided to the general public.

Further info can be found by searching on "near zero volt storage". Due to recent Li-ion transportation restrictions there has been a flurry of research on possible ways to improve Li-ion cells so that they can be safely stored at near zero voltage (near 0V) because this greatly increases safety, e.g. it increases thermal runaway threshold temperature, and greatly decreases amount of gas production, and eliminates all toxic gases, etc. Many of these recent papers link to the (scarce) earlier literature on overdischarge topics, and give brief overviews.

For example, see Kyle R. Crompton's 2017 RIT Thesis Enhancing Near Zero Volt Storage Tolerance of Lithium-ion Batteries. Below is an excerpt from pp. 19-20.In the following section he mentions some prior research on related matters, which might be a good place to begin any further investigation on this topic.

KyleR.Crompton said:

It is important to note here that determined values of the onset potential for copper dissolution can vary since the choice of an oxidative current density threshold is arbitrary and non-faradaic process can contribute to the oxidative current. Additionally, the exact value of the onset of copper dissolution can be affected by factors such as ambient conditions and electrolyte composition. In the present work, 3.1 V vs. Li/Li+ will be assigned for interpretation purposes as below this value, no oxidative current was observed in the linear sweep voltammogram in Figure 9d, and therefore, no copper dissolution is expected to be occurring.

The adverse effects from copper ions being present in the electrolyte will lead to side reactions with the electrolyte, competitive reduction processes with lithium ions, and copper dendrite formation which can lead to internal shorting. Each of these effects can cause significant damage to the cell and reduce its recharge capacity and performance.[8591]. Thus, for cells to be tolerant to near zero volt storage, cell design must be modified or different materials used to avoid the copper dissolution degradation mechanism.(emphasis in original)

Click to expand...

---

Below are excerpts of Samsung datasheets showing the 1.0V lower prechargelimit, from Dec. 2017and July 2015.

---

I don't try to recover 0V cells. They never turn out to be any good nor worth the effort. It goes directly into the bin.

However I'd say about 30% of the cells that I process are under 1V and they usually recover well. And I've processed thousands of these cells. I usually just put them together with other weak cells in a parallel battery tray of 8 cells. That's my current setup right now and I charge it up with an imax clone. Usually flip to NimH setting at about 2A and let it charge up slowly. That will spread the charge amongst 8 cells. It usually takes only a couple minutes and then I flip back to the lithium setting once it reaches enough voltage for it to accept a charge. You do want to check on on heaters after a while. Nothing too complicated.

These cells have been in service for two years now in my powerwall so there isn't any ill effects yet. None of my packs have suffered any degradation yet since they all stay balanced. So it's been fine for me to process these under 1V.

^^^ Sigh, your're greatly increasing your risk by using cells that were below 1.5V. They may seem fine for months or years as the bomb slowly ticks while the dendrites slowly grow during cycling. One they grow large enough to pierce the separator they can cause internal shorts that can lead to uncontrolled thermal runaway and possible explosions and fires. It's not worth the risk.

Thank you for your responses.

^^^ The problem is that the number of people that looks for cells (and starves resources without the need from themselves) had increased so much that you can only get what others left behind.
I came to scavenging after a fiend sent me 1000 units of heaters, and 200 units drilled after paying for 1400 checked and selected ones.

---

floydR said:

if single cells/cells in parallel
imax 1s mode
icharger X6 1s mode
icharger x8 1s mode
if cells in series
up to 6s imax, icharger X6
up to 8s Icharger X8 I recovered 6 24v 7s batteries with my x8 used .02Ato charge to 21v then 2A to24v then 4A to 28.5v
later floyd

Click to expand...

Floyd, and those devices you can also use as pack chargers/manual leveling?

Yes the imax and x6 up to 6s. The x8 up to 8s. But beware that low voltage cells<1.5v have a potential for developing an internal short/fire risk. as detailed in post 6 of this thread.
later floyd

I added to post #6 some excerpts of Samsung datasheets which warn not to charge cells below 1.0V.

You Will soon see that the time doing below 1v cells yields so low outcome of ok cells that its not worth it. I only do cells that the Opus Will allow. Thats 1v limit. Cells below 2v sitting for some time perhaps yields 20% ok cell max.

IF a cell tests fine for capacity and self discharge i use Them. But they only get One Chance.

daromer said:

[...] IF a cell tests fine for capacity and self discharge i use Them. But they only get One Chance.

Click to expand...

Worth emphasis: an extremely overdischarged cell may test fine for capacity, IR, etc,but this does not imply that it is safe to use. The problem with dendrites is that these cellular tumors typically don't produce any characteristic symptomsuntil it is much too late and - due to cycling - a dendrite has grown long enough to pierce the separator - causing an internal short. The timescale for sucha growth process is unpredictable - it may take days, weeks, months, years (or never happen).

Its true. As of today date There Arent any Good pappers for it either. Some exist but not many where they actually done proper xrays.

IF you wont take the risk dont do it. Dont even do DIY

My Wall is now 5 years and Still No signs. But i also built it upon basic rulesets and understanding. Thats not a say that it wont crash tomorrow.

^^^ If your OPUS routine rejected cells below 1V (as you seem to imply above) then likely you should not have to worry about shorts due to extreme overdischarge. Congratulations on your 5 year success.

Does the <1v cutotff apply to all cell form factors? Another member and I each picked up a LG RESU HV with< 1 volt cells.
I lucked out and got the better of the two units. With 4 of the 6 7s batteries sitting above 7v, I think 1 cell was below 1v out of 28cells the other 2 7s batteries may have had more but charged .02 amps u to 21v. Same as I charged the other batteries. kinda worried as I am going to use the 2 batteries that are @24v to do a regenerative capacity test the other 6, 1 at a time of course.
later floyd
This what I know about the cells: 63Ah lgx JH3 7s wiring harness for each battery

floydR said:

Does the <1v cutotff apply to all cell form factors? [...]

Click to expand...

Unfortunately generally the manufacturer's don't publicly release this information, so what little we have to go by is what is leaked (e.g. the Samsungdatasheets I quoted in post #6). Iirc some manufacturers specify a 1.5V lower limit but I didn't have any luck finding that now by asearch.

On one hand, we know from general theory that copper can dissolve at low voltages then - in subsequent charges- nonuniformly replate into dendrites. Otoh, perusing patents shows that there have been some improvements to help counteract such(e.g. chemistry tweaks, separator improvements, etc).But we don't know which cell models have these improvements, norhow much theyimprove the lower voltage threshold. However we can infer that it is not thecase that all 0V cells are safe, else there would be no need for Samsung's warnings, nor any need forall the current research in near-zero-volt storage (cf. post #6).

Thank you, I suspected lithium polymer batteries haven't had as much released research as 18650's have.

later floyd

^^^ What I quoted above likely also applies to most commercially available LiPo (pouch) cells too, since either they use liquid electrolyte (same as an 18650), or a gel polymer electrolyte (not solid electrolytes - which are rarely used commercially). It's not clear if gel polymers will perform any better than liquid electrolytes to inhibit such dendrite growth. Alas, there seems to be little literature on that topic.

I intend to recover only cells above 1,5v, The ones below will probably go to "experimentation". There are only 20 or so in 2000.

I've ordered some parts to build a normalvoltages charger, tp4056 based, and then a low voltage recover rig.
I'll add temp probes to the cell holders and using some multiplexers I have around, monitor temperature and voltage for each cell with an arduino. I'll try to make the arduino send the values via mqtt and stop charging if something goes wrong.

Something like eight cells in parallel charged using a single TP4056 with resistor adjusted for 400mAh. But I still worry about some things:
If I was to put eight cells in parallel they always should have the same voltage or they may suffer while self leveling alone.
If a single cell is disconnected from charging, that would raise the charge current for the rest of them.

Modding a full batch of TPs for 50mAh seems to me... drastic at least. I suppose it could be done, though. Perhaps removing the stock resistor and piggybacking a small board with several resistor values to be able to select current.

Do you find possible to only use one set of resistors for all the TPs and bridge all to it? Perhaps that resistor is tied to ground that can be common.