What's the longest period of time you can store a Li-Ion pack?

Hi ESFMs,

I recently had to swap out one of my packs on my bike due to issues I'll be diagnosing later.

The pack I just swapped in is a 48v 20Ah and I only used it for around 5-6 months in it's previous life. I put it away at the time because I was upgrading to a 30Ah.

At the time it was in excellent working order and had it's full range. But now, it doesn't charge all the way full (stops charging at around 52.6v) and it starts shutting down as early as at around 49v. I can unplug and plug back in to get it going again but then I have to feather it to avoid shutdowns, which end up happening anyway. Makes my ride an infuriating one.

Anyway, since I stored this pack for over 3 years, it's clear that maybe it's degraded. I think I might have stored it fully charged for some of that time but later on, I made a point to drain down to half. But it might have been too late when I did it sometime last year.

Does this check out? Is this what you can expect from over storage of a pack or does it have a different issue?

Also, and mainly, how long can you get away with storing a pack before it's no good no more?

Its frustrating because it makes keeping a spare almost unfeasible if they're gonna die in storage.

Clinton

Do you know what particular cell was used in this pack?

Pajda said:

Do you know what particular cell was used in this pack?

Click to expand...

I have no idea. I bought this pack from an eBay vendor in Australia (I'm in Perth western Australia myself). I can't find the listing in my purchase history. It only goes back to 2017. I bought it in 2016. I don't remember the name of the vendor, otherwise I'd shoot 'em an email. They're definitely the 18650 types. That much I know. I'll do some digging.

Quite a few threads on batteries here. There are a few rules to follow if you want max life but all batteries degrade over time.
This is what I follow:
Don't run them down to low, Don't charge them up to 100% and not use them. don't let them stay to hot or cold.

Have seen charts that some batteries will lose 5-10% capacity after only a few months of setting fully charged with out use. The loss drops way down below 80%.

Also, and this should be #1: if you can't charge the pack occasionally, disconnect the BMS! If left connected it will slowly drain the pack to power itself.

ZeroEm said:

Quite a few threads on batteries here. There are a few rules to follow if you want max life but all batteries degrade over time.
This is what I follow:
Don't run them down to low, Don't charge them up to 100% and not use them. don't let them stay to hot or cold.

Have seen charts that some batteries will lose 5-10% capacity after only a few months of setting fully charged with out use. The loss drops way down below 80%.

Click to expand...

No, my issue is how to store a battery you're not using for a while. This is a pack going unused for 3 years.

See if you can check each cell voltage with some sort of meter. The pack sounds like it went out of balance. Just get it back in balance and it should be good. But since it drained without use, there might be a weak cell in the pack. A weak cell will give you balancing problems and also slow drain. But if you use the pack everyday you won't notice it too much.

I been using these tenergy cell meters on my packs that I don't use for a while. Its lcd so it doesnt use much power. It sounds an alarm if cell voltage goes too low. I lost some packs that drain all the way down from sitting too long, now at least I hear the alarm and charge them back up. It save me a couple times already. For your large pack you would need several of these.

I store my packs fully charge and never had problems, except with the ones that had a slow drain. But they are suppose to last longer if you don't store fully charge.

You need to know the state of charge that means what are the voltage is of each parallel group of cells. You must write it down on a piece of paper no not in your mind but on a piece of paper like.
1. 3.90v
2. 4.01v
3. 3.85v



13. Xxx volt
The voltage of the battery
The voltage of your battery charger.
This is where you start from.

With the original question in mind.... Storage of packs is dependant on a few things.
Cell quality aside (hopefully you can find out what's they are) the main factors for cell degradation are SOC, and temp. So assuming good quality cells...
Optimal conditions are cool/cold ( not freezing) and 40-50%SOC with periodic balance to compensate for bms draw. Then 2years would be doable, with some minimal damage, and it would probably seem to function fine. Metering would reveal some change in capacity and IR.
however, if you put it away fully charged and in a hot place, I would expect noticeable loss in 6mths.
Got a discharger? Measure it and tell us. 3yrs is a long time.
I've just restored a pack (LG cells) after about 3yrs of dormancy. Stored at 50% in a variable but usually cool shed. Seems to be working well, but I haven't measured anything sorry. It needed a balance bc the bms draws of the first group only but apart from that it's still pumping out1kw pretty well. Not too much sag. Capacity is probably down and ir will be up but I expect that.

LeftieBiker said:

Also, and this should be #1: if you can't charge the pack occasionally, disconnect the BMS! If left connected it will slowly drain the pack to power itself.

Click to expand...

Very good point. The BMS is likely the culprit. I'm not saying the BMS is bad, but you have to realise that BMSs have a very small but very real power consumption (quiesent load, usually in the order of 5 to 40 uA). It very very slowly drains the battery (in months to a few years depending of battery state of charge). The other thing to realise, is that quiescent load is not pulled from all the cells (load not distributed equally on all cells). Typically, the first to first three cells goups (B1 or B1 to B3) on the negative side of the battery will be the ones from which the BMS pulls the load. So for instance in a 14S battery, while most cells will have not changed in voltage very much if at all, cells of groups B1 (or B1-B2, or B1-B2-B3) will have gone much at lower voltages compared to the rest of the cells in yhe other groups (BTW group B1 is the closest to the negative batt terminal). The best way to diagnosis this is to measure voltage of each different goups in series (aka each B group) individually. The best way to fix the unbalanced problem is to slowly charge (at 1A) each individual B-group individually back to 4.20V one at time (a process called manual balancing) by charging through the BMS balance wires.

Realise that when one cell group is unbalanced relative to all the others cell groups of the battery, it triggers the BMS and reduces the overall battery capacity artificially. When you bulk charge (without balancing), most of the cells (ideally all the cells) will reach 4.20V or so and the BMS will then trigger high voltage cutoff (HVC) even though group B1 and B2 might still only be at 3.85V (because BMS drained them more from storage and quiescent current). Thus your battery does not appear to hold a full charge anymore as overall battery voltage is lowered from B1 and B2 having lower voltage.

On the flip side, once you discharge your battery (riding your bike) once B1 and B2 hit the low voltage cutoff (LVC) of say 3.25V then the BMS shuts the whole battery OFF even though groups B3, B4, B5, B6, B7, B8, B9, B10, B11, B12, B13 and B14 are probably still all at 3.40V with some juice left. Thus even though your battery has an overall voltage that still indicates it has some juice, the lower cell groups B1 and B2 might prevent that same juice from being used because the BMS is trying to protect against overdischarge of B1 and B2 by triggering LVC (the BMS senses some B1 and B2 hit 3.25V and stops the whole battery).

Last thing if your have a "balancing" BMS (typical balance current is usually 40-60 mA only) it might take days or weeks to rebalance the battery by letting it on the charger and let the BMS re-equilibrate everything.

But realise a lot of battery BMS out there are not real "Balancing" BMSs but rather just protective PCB (with preset LVC and HVC protective values and overcurrent protection) but no real "balancing" function. In that case the only way to recover your battery is to manually balance it (get yourself a genuine SKY RC Imax B6 to do it) through the BMS sensing leads or it's JST connector. And always remember to disconnect the BMS connector (the JST connector) form the battery before long storage periods to stop parasitic quiescent current from draining your battery because of the BMS quiescent load. Store at around 30-50% to avoid premature cell aging, in a cool and dry place. If your BMS is unplugged, you can even store at 20% as there is no load on cells and can live without the fears that the cells will never be discharged lower than the fatal 2.5-3.0V at which battery destruction starts to occur.

Hope this helps,

Matador.

So if for a 48V batt (13S) if t should charge to 54.6V, but yours now stops a 52.6V. That's 2.0V lower than normal.
And the battery shuts of at 49V but i'm thinking it should normally shut off at around 39V (LVC 3.00V per cell).

I'm thinking your battery might look something like this:
When CHARGED --> DISHCHARGED
B1: (-) 3.29V --> 3.00V (LVC trigger)
B2: 3.29V --> 3.00V
B3: 4.20V --> 3.91V
B4: 4.20V --> 3.91V
B5: 4.20V --> 3.91V
B6: 4.20V --> 3.91V
B7: 4.20V --> 3.91V
B8: 4.20V --> 3.91V
B9: 4.20V --> 3.91V
B10: 4.20V --> 3.91V
B11: 4.20V --> 3.91V
B12: 4.20V --> 3.91V
B13: (+) 4.20V --> 3.91V
Total: 52.8V --> 49.6V

So probably you need to first bulk charge the whole battery full, then charge B1 to 4.20V via the (-) (aka B0) and the B1 balance lead, but check voltage first to know what you are charging (if its the right cell group). Probably need to balance B2 also through balance lead B1 and B2. In any case, be very mindefull of polarity (check voltage first!), Otherwise is you charge through BMS lead and reverse polarity, you might fry you BMS.

Matador

Storage of **cells** is one thing, and does vary by model, chemistry, temperature etc

this means fully isolated, nothing connected, only self-discharge rates to compensate for, checking and topping up monthly should be fine, even less as you become more certain voltage per cell gets nowhere near say 3V.

however if circuitry included in a pack, e.g. BMS, is left connected then all bets are off, if you can't isolate the cells I would start off checking even weekly

This is a pretty classic case of BMS quiescent current unequally discharging only certain cells in the pack and the resulting inbalance triggering LVC too early when discharging and triggering HVC to early when charging. I'd say if no cells went under 2.5V, you have a 95% chance of being able to easily fix it by measuring individual cells groups voltages and manually charging the lowest groups back to 4.20V to bring them back to the same 4.20V of the other normally charged groups.

Matador.

Let us know how it goes !
Matador.

Thank y'all. That answers a lot. And thank you Matador for that in-depth explanation. I have some other packs to go save by disconnecting their balance connectors.

It was shutting down even sooner today. And this time, if I simple unplug and plug back in when it does it, it doesn't always work. I have to wait.

I'm lucky that I ride with 3 packs on one bike. So I can switch pack from time to time.

Looks like I have 2 packs to fix (well 3 actually, I have one with a damaged cell as well)

Matador said:

This is a pretty classic case of BMS quiescent current unequally discharging only certain cells in the pack and the resulting inbalance triggering LVC too early when discharging and triggering HVC to early when charging. I'd say if no cells went under 2.5V, you have a 95% chance of being able to easily fix it by measuring individual cells groups voltages and manually charging the lowest group back to 4.20V to bring them back to the same 4.20V of the other normally charged groups.

Matador.

Click to expand...

How can I manually charge individual cells like that? What do I need? What's the procedure?

ClintBX said:

How can I manually charge individual cells like that? What do I need? What's the procedure?

Click to expand...

Here is a video example I found on youtube (he does it with an unbalanced lipo pack... Same prcedure with an ebike battery. Just unplug BMS first by pulling the JST connector ( the heavy gauge battery main wire that carry all the current can stay plugged to BMS, no need to remove the too).

[youtube]DP1-VBNJmsg[/youtube]

I use personnaly use a genuine SKYRC Imax B6AC v2.0 to do this very same procedurr. Works like a charm. I made an adaptor with PCB board pin connector and a wire to plug, probe, and charge through JST BMS wires connected to the battery.

Matador.

Be very carefull to not short anything.
Be very carefull to plug in the right holes in the JST connector (don't skip one hole by mistake).
Be very carefull not to reverse polarity when tapping in the JST connector.

Take your time, don't rush anything.
Probe voltage first (don't charge anything before first checking polarity/voltage) to assess cells and to make sure you know the polarity (what is + and what is - on that JST connector).
Use color codes (red+, black-): use a red marker to put a color code on the (+) end of the JST connector and a black color code on the (-) end of the JST connector. Mark your wires red (+) and black (-) too (i used colored heatshrink).

Matador

ClintBX said:

Anyway, since I stored this pack for over 3 years, it's clear that maybe it's degraded. I think I might have stored it fully charged for some of that time but later on, I made a point to drain down to half. But it might have been too late when I did it sometime last year.

Does this check out? Is this what you can expect from over storage of a pack or does it have a different issue?

Click to expand...

As I explained in my last messages the problem is most likely to be related to BMS have unbalanced cells over time in storage.

As for knowing if your cells have degraded. Well there have been some scientific studies on this phenomenon called "calendar aging of lithium cells". Here is the scientific paper with results of the studies:
https://endless-sphere.com/forums/viewtopic.php?f=14&t=86777&hilit=Calendar#p1269133

But in a pinch, lets say you stored your battery fully charged at 100% state of charge (SOC 100%) during 3 years at 25 degree celcius. And lets say your 18650 cells are NMC chemistry (most likely), then the first year your cell capacity will have deteriorates to 94% (18.8Ah), after two year you should be at 88.36% (17.7 Ah), and after three year you should be at 83.0584% (16.6Ah).

Still should be plenty usable !

If you had left it three years at 50% SOC instead of fully charged (at 100% SOC), then after one year, residual capacity would be at 97% (19.4 Ah). After two years, residual capacity would be 94.09% (18.8 Ah). After three years we are talking around 91.2673% (18.2 Ah).

So really, i don't think cell degradation is the issue here. The BMS current load and resulting voltage cell imbalance created is the issue here.

Matador.

I agree with Matador, a poor BMS design he describes causes a lot of issues and I guess that this is also that case.

Just remark to the cell ageing itself:

The recent tests (for example docware`s work here at ES) proven that modern HE (high energy density) cells, shows significantly better cycle life than HP/HD (high power density) cells with no general connection to actual cell chemistry. Based on these results, it can be assumed that a similar phenomenon can occur in the calendar life as well. So that was reason of my question about used cells in the pack.

Because available scientific tests are in the most cases misleading in my eyes, because they used perfect scientific methods but unfortunatelly on the "most shittiest" cells ever produced in most cases. I started a specific calendar ageing test for the "most perspective" cells based on my and docware`s cycle life results. Those are 29E7, M29, M36 and 50E. Test is conducted for 100% and 50% SoC storage at the "shelf" temperature. Now after 5 months the preliminary results shows that there is a measurable calendar wear but it should be rated as rather small, and maybe most interestesting result is that the absolute difference between storage at 100% and 50% SoC is below 1% for all four cell types. I will soon add up to four more cell types to the test with the focus on modern HP cells. I think it will be 30Q and VTC6A, where I am expecting significantly worse calendar ageing - similar to the test results posted by Matador.

I agree with you. It seems more modern cells age better than the ones used in the scientific paper on calendar cell aging.
I personally have harvested 218 Sony VTC4 18650 cells (and 68 Sony VT4 cells) from Makita drill battery packs.
I tested each one of these cells individually for capacity.
With the date of manufacture code written on each cell, I was able to see the exact date of manufacture of each cells.
So I know how each cells are old.

I have recorded the results in an excel file.
Post : https://endless-sphere.com/forums/viewtopic.php?f=14&t=28285&hilit=VTC4&start=175#p1299346
Excel file: https://endless-sphere.com/forums/download/file.php?id=217443

On average, the 218 VTC4 cells were 2.4 years old (Standard deviation 0.5 yrs).
On average, the 218 VTC4 cells were 2190 mAh capacity (Standard deviation 52 mAh).
I bought them from DoctorBass who stored them in his garage, so I'm guessing average storing temperature was between 20°C.
I also measured an average of 21.8 milliohm internal resistance (DCIR) (standard deviation 1.5 milliohm)
Storage voltage ranged between 3.56 and 4.01 volt (measured when I salvaged the cells).

But still the VTC4 are rated at 2100 mAh, and I measured 2190 mAh. So they were around 100% capacity after 2.4 years.
And these are high power density cells (30A cells)!

Matador

ClintBX said:

I bought this pack from an eBay vendor in Australia (I'm in Perth western Australia myself). They're definitely the 18650 types. That much I know.

Click to expand...

To answer your original question a pack with cells in relatively good condition (whatever that means) with let's say no more than 50mv variance from the highest to lowest cell (e.g. 3.852v to 3.857v) when stored should still be in balance without a self-discharge voltage [for the sake of this discussion] no more than (e.g. 3.750v to 3.850v), if you get my drift. NOW that's assuming the pack hasn't been abused (whatever that means). For the sake of this discussion, let's say no more than 100-200 charge/discharge cycles. From what you've posted so far there are just too many variables to answer your question without more specific info. Your BMS may not be the problem, but rather the condition (variance in voltage and IR) of all those 18650 cells.

Will tell you that I had two like new Parkzone 2200mAh Lipo batteries [very similar to Lilo] stored for seven years at 3.85v. This year I decided to parallel them together (4400mAh) and still charge them at 3S with my Balance Charger. Both packs were still at 3.85v before I paralleled them together... unbelievable but true. Also 5 of the 6 cells i tested before paralleling had the same IR of just 3 with the other cell having an IR of 4. I tested the IR on a couple other Lipo 3S packs one of which was slightly puffed and it's 3 cells averaged an IR of 33. Typical of a puffed (abused) pack.

Bottomline: There isn't a whole lot a BMS can do other than act as a safety protection (BPS).

Matador thanks for your comment.

I'm trying to warn against generalizations of evaluation of the cells by some significant parameters like chemistry, and right now I am doing the same thing. Both me and docware measured excellent cycle life results of LG Chem HG2 which is definitely a contender of HP/HD cell category, while VTC6A and particularly 30Q shows poor results. So I have no problem to believe that VTC4 should be also excellent cell in all parameters.

Also it is possible to find a general result that calendar life is not a significant problem for majority of modern cells at all.

eMark said:

ClintBX said:

I bought this pack from an eBay vendor in Australia (I'm in Perth western Australia myself). They're definitely the 18650 types. That much I know.

Click to expand...

To answer your original question a pack with cells in relatively good condition (whatever that means) with let's say no more than 50mv variance from the highest to lowest cell (e.g. 3.852v to 3.857v) when stored should still be in balance without a self-discharge voltage [for the sake of this discussion] no more than (e.g. 3.750v to 3.850v), if you get my drift. NOW that's assuming the pack hasn't been abused (whatever that means). For the sake of this discussion, let's say no more than 100-200 charge/discharge cycles. From what you've posted so far there are just too many variables to answer your question without more specific info. Your BMS may not be the problem, but rather the condition (variance in voltage and IR) of all those 18650 cells.

Will tell you that I had two like new Parkzone 2200mAh Lipo batteries [very similar to Lilo] stored for seven years at 3.85v. This year I decided to parallel them together (4400mAh) and still charge them at 3S with my Balance Charger. Both packs were still at 3.85v before I paralleled them together... unbelievable but true. Also 5 of the 6 cells i tested before paralleling had the same IR of just 3 with the other cell having an IR of 4. I tested the IR on a couple other Lipo 3S packs one of which was slightly puffed and it's 3 cells averaged an IR of 33. Typical of a puffed (abused) pack.

Bottomline: There isn't a whole lot a BMS can do other than act as a safety protection (BPS).

Click to expand...

It's possible some cells may be in bad condition (higher DCIR, self discharge...), but I assumed since it had only been used for 6 months that it probably had a low number of charge/discharge cycles. And since it's a pretty big pack (20Ah), less likely to be abused than, say a 10Ah pack which would likely have a lower number of cells (and lower max discharge rate).

That's my rationnale for thinking that the BMS (and it's quiescent current on row B1 or B1-B2 or B1-B2-B3) is the likely culprit for throwing the overall pack off balance making it a poor performer after 3 years storage.

Not impossible some cells have gone bad. But I think it's more likely the pack is out of balance from BMS parasitic drain. Think about it, a BMS with a 100 uA parasitic drain, stored for 3 years (3 x 365 day x 24h/days) that's still 2.628Ah drained off of one row of cells (enough to through the battery off balance). And if the pack performed great before storage but now poorly after storage, it's likely due to the BMS draining cells during that time where the battery pack was not particularly stressed.

That said, nothing is impossible, but it'd be good to try to revive it!

Matador

Pajda said:

... while VTC6A and particularly 30Q shows poor results.

Click to expand...

There's about 75 cents difference between the HG2 ($4.50) and the newer 30Q 141 ($3.75) in quantities of at least 100 ... depending on supplier. With the old saying that you get what you pay the HG2, still looks to be the better buy in the long run ... depending of course on application.

At this point in comparing prices it's possible LG Chem has raised their price because of the previous bad publicity with 30Q's reported self-discharge tendency. Samsung may have decided it has to lower it's newer 30Q 141 price in order to keep potential clients from jumping ship to HG2. For those that want to get the most bank for their buck the HG2 20A probably still has a leg up on the 30Q 15A even though some say it should be rated closer to 20A. Voltaplex shows the HG2 as one of their Hot sellers ... https://voltaplex.com/lithium-ion-battery/18650-battery?is_ajax=1&p=1 ... whereas 30Q isn't a hot seller.

I've got over 100 charge/discharge cycles on my 10S3P experimental pack of 30Q 141 cells. So far no self-discharge problem as previously reported with the 136 cells. So, hopefully Samsung solved that reported self-discharge problem ... time will tell as its in storage over the winter from mid-November until mid-March.