Are my Li-Mn batteries toast?

I have a pair of 20ah 36v LiMn batteries that were originally built by electricrider.com in 2012 for a 72v/40a crystalite system. I believe they are sanyo cells..

I went for a ride the other day and they performed well. I did not immedeatly charge when i got home.. today i went to hook up my chargers and got nothing.

Volt meter reports 0volts on battery A and battery B was at ~16v, then that voltage rapidly began to fall within a few seconds of the volt meter being connected.

Not sure how many total charge cycles i have on them but I commuted ~30 miles every summer day for 2-3 years & since that time, just a few rides here and there, so maybe a few hundred cycles. They've been working well for the past several months and are now completely drained & not taking a charge.

Are these batteries toast? is there hope to revive them?

If yes, what is my best course of action for success?

if no, can anyone recommend a viable replacement with similar performance?

Thanks in advance

Welcome to the forum.

7 years is a good long life for a Lithium battery, and It sounds like a bad battery, but it could also be a failed BMS. It may be worth your time to open up the battery and look for voltage on individual cell groups. If you have voltage on the cells but not on the main battery leads, then the BMS has shut the battery down. If it has, there may be a good reason, it it might have failed. you'll need to look at the voltage on each cell group. If any of them are below 2.5 volts, then thank the BMS for doing it's job and look for a new battery.

Thanks for the reply...

For what it's worth, & as far as I know, these packs do not have a BMS.. The LiMn batteries were supposed to be able to handle loads in excess of 100A (unlike comperable LiFe batteries available at the time) & thus the BMS was not integral to the pack's design.. or so I was informed by electricrider.com circa 2012

That said, at this point I have not opened the packs, so I cant be certain there is no BMS.. I will do what you suggest and crack these open to test the individual cells, as I have nothing to lose at this point.

Thanks again!

Sounds like a flaky connector to me.

It's definitely not the connectors.. using xt90s connectors that I soldered myself and continuity is good across all of them.

Even if hook the multi-meter to bare wire the result is the same.

Further, i have two 36v/20ah packs & 2 chargers behaving similarly, so no one connector could possibly be the issue.

My thought is that the packs are generally worn from use over the years, PLUS, I left them in a state of deep discharge for several days & did not disconnect my 5v(USB) & 8v(headlights) bucking converters, which allowed the packs to be drained below the low-voltage cutoff...aaaaand pOOf!

Still interested to hear from anyone with experience using these Sanyo LiMn cells.. Can they be revived after deep discharge?

Thanks!

Sorry I missed that the pack was deeply discharged. LiMn cells are very forgiving. I'd suggest opening the pack to see if there is a BMS and if so bypassing it to get it at least recharged to the point where the BMS can be used again. The pack may not perform as well as it did.

ok.. I havent fully disassembled the pack yet, just peeled back some of the outer casing, & it does not look like there is a BMS..

there is a label that says:
"Max discharge 50a (limited by automatic reset braker"
"Minimum Voltage 31v (External low voltage cutoff required)"

I can see the 50a braker & it appears there is no other pcb circuitry or BMS.. & i assume any low voltage cutoff was handled by my speed controller.

Since my charger isnt charging & it appears there is no bms, how do you suggest I apply a charge to this thing to get it back up above the minimum voltage?

Thanks again



ps- spoke to electroride.com & he congratulated me on having my batteries last this long.. unfortunately its just hard to stomach that they are both completely dead & unusable after performing so well just the other day... ‾\_(ツ)_/‾

Keep peeling! The "Braker" sounds like a BMS.

You are playing with fire! The cells are already damaged and will short internally (or grow existing very small shorts). LiMn has good longevity but is still a reactive lithium chemistry. It can go up in flames.

Nope, not seeing a BMS at all, just a 50a in-line circuit braker... seeing the same voltage (~19v) on either side of the braker (at bare wire coming out of the pack as well as at the xt90s power connector

checking the wayback machine for electricrider.com circa 2012, their older LiMnO2 batteries & their manuals make no reference to BMS anywhere. Fast-forward to 2015 & they were offering lifepo4 packs which do explicitly state a BMS is included..

so, based on my first hand observation and other research, I presume my packs have no BMS


oddly, wen i checked the voltage on this pack earlier, it was at 16v, now its at 19v... huh?

Cells can "bounce back" after a load is removed, and some may take some time to do so.

Voltmeter batteries, as they get lower, also cause the meter to read a higher voltage than is actually there.

FWIW, lithium cells that have actually drained to zero (really, below the cell's specification for LVC), have physical damage inside that's unpredictable in it's results. LiCo cells are usually more problematic than other types, but it applies to most of the Li chemistries.

Once the damage occurs, it's possible for a random event to occur inside the cell, at any time, charging, discharging, or just sitting there, that causes rapid heating and potentially fire (even explosion in an 18650 or other can type cell). Becuase of the differences in construction and constraint of the cell interior, this kind of failure is probably more of a problem in can-type cells than pouches, but it could happen in either type.

The higher the currents (and thus internal heating), the more likely the event probability is, whether in charge or discharge.

So while you *can* sometiems recover dead cells and reuse them, I highly recommend never bringing them inside a building, or keeping them near enough to one that a fire in the pack could set fire to the building.


If you do want to attempt to recover it, I do not recommend bulk charging until it is at a "normal LVC" point or above, with all cells at the same voltage.

Use a regulated adjustable-current adjustable-voltage power supply (lab psu), and set the current for a few mA, and the voltage for the cell's normal full-charge voltage or less.

Connect it to the most negative cell group first, across the + / - of the first cell in the group, and monitor it's voltage and temperature. The current is so low that temperature should not change from ambient. A cell warmer than that indicates a possible problem.

If you like, you cna up the current to a few dozen mA to maybe a few hundred mA, *after* the cell group has reached at least it's normal LVC point, or above.

ONce that cell has reached the voltage you set the psu to, then disconnect both leads, and move htem to the + / - of the next more positive group of cells.

Continue this process until all cells are at equal voltage.

If you find any cells are warming up in this process, do not use them; they may have failed internally, and most likely the group they're in is not charging to the set voltage, and/or is draining from that voltage just sitting there.

amberwolf said:

...

Click to expand...

ok.. let me just say that this was, by far, the most comprehensive & useful response to any forum post I've ever seen, ever.

Thank you Amberwolf.

As I lack a proper lab PSU, and these packs contain 130 can cells each, which are all spot-welded.. i dont think i'll be using this recovery technique... to time and labor intensive, not to mention risky.

Perhaps in a few days they will bounce back to a chargeable voltage?.. if not, i think my time and money will be better spent on new batteries alltogether...

ideally i want 20ah battery that i can use with my existing 72v/40a crystalite system.

Electroride.com is advertising LiNMC as their top of the line battery & I see GRIN is selling LiGO... which is the preferred? & what are the benefits & drawbacks of each..

Also, does anyone have a use for these old packs? or can anyone recommend an adequate method of recycling?

Thanks again, everyone.

Scratching my head for a while, I seem to remember Koinion cells were though to be "self balancing". They aren't really, but they had really bad charging efficiency at their top end, so they tended to stay more balanced than other cells, back in the lithium dark ages of 2007. That might be what your pack is, and why there is no BMS. They were LiMn cells, with something different about them. Modern knowledge is that all lithium needs some form of Balancing, either with a balance charger, or onboard BMS.

The LiGo packs are awesome, but may not be the best for what you want. they are made with LG MG1, LiNMC cells, and while being very good for what GRIN's goals were for the use of those packs, may not be the best cells for your needs.

Before we get too far, I got a math question for you. You said you had Two 36v 20Ah packs, but you also said you were using them as a 72v 40Ah? They don't add up that way, but i saw you post it twice, so I'm curious which it is? two 36v 20Ah packs stacked in series would be 72v 20Ah, while two 36v 20AH set in parallel would be 36v 40Ah. both equal 1440 watt hours.. 72v 40ah would be 2880 watt hours, 4 of those 36v packs.

The LiGo packs built into a 72v20AH pack would take 16 of them @ $2240. or for a 72v 40ah pack, 32 of them, @ $4480. It would be an awesome battery, but massively overkill. These are packs built to be safe and portable, designed so you can carry them legally on an airplane, for things like e-skateboards, or small folding bikes.
A name brand cell pack from a trusted vendor like EM3ev.com or Lunacycle.com would run less than half that.
for examples: https://em3ev.com/shop/em3ev-72v-20s10p-rectangle-battery-pack-24-5ah-29-5ah/
https://lunacycle.com/72v/

joeuser999 said:

ok.. let me just say that this was, by far, the most comprehensive & useful response to any forum post I've ever seen, ever.

Thank you Amberwolf.

Click to expand...

I'm good at "comprehensive"....useful depends on the point of view. :lol:

As I lack a proper lab PSU, and these packs contain 130 can cells each, which are all spot-welded.. i dont think i'll be using this recovery technique... to time and labor intensive, not to mention risky.

Click to expand...

Well, you don't have to disassemble the pack to do this--just connect the PSU across each parallel group of cells (10 at a time, I'd guess, in 13 groups).

It is definitely time intensive, and can be labor intensive.

My own reason for not reusing such a pack would be safety rather than time or labor, as I've read of too many fires of unknown cause that probably had to do with various versions of this type of problem (because at least some of the packs were either made of recycled cells (tested or untested) by that person, or were bought so cheaply as "new" packs from China that they were probably recycled cells of unknown condition...).

Perhaps in a few days they will bounce back to a chargeable voltage?..

Click to expand...

Unlikely. Even if they do, the same safety issues still apply.

And when cells that have gone this low don't actually fail, they're still less capable than they were--higher internal resistance means they can't discharge as high a current as they could before, and they['ll also have lower capacity.

Electroride.com is advertising LiNMC as their top of the line battery & I see GRIN is selling LiGO... which is the preferred? & what are the benefits & drawbacks of each..

Click to expand...

LiGo is just LiCo cells in 10s1p modules potted for weatherproofing/etc, with it's own BMS. You'd have to parallel several seriesed pairs of these to get the pack size you're after. But if one ever failed it'd be easy to replace out of the pack. THe BMSs would also protect each individual module.

Disadvantage of that is if any cell goes low, it cuts out that module and places a higher load on the remaining modules, where in a fully-paralleled system (like the packs you have now probably are), all cells in a parallel group "help out" all the others, instead of being isolated from each other like the LiGo packs are.

Advantage of that is one bad cell in a "group" of LiGo packs can't take out it's parallel neighbors, which can be a problem in a fully-paralleled pack.

NMC is a good chemistry, but it doesn't say anything about what the cells actually are; what they're capable of, etc. You'd have to get the brand and model of cell from the seller, and look up what hte manufacturer says, to find out the actual specs (which would then usually be derated some for use inside a pack, for various reasons). (unless you trust the pack seller to give accurate information). Couldn't say what drawbacks or benefits their particular pack might have.

I use NMC cells on my trike, but they are used pouch cells from EIG, 20Ah each, like those that Zero used on their motorcycles several years back (don't remember which year). I've actually recovered three of them from "dead" and used them as a small lighting pack fro quite a while until again they got run down too far, and one of them actually reversed voltage and puffed up. That set is all retired now. But they're hard to find, and expensive when people do sell them, even used. :/

Drunkskunk said:

Before we get too far, I got a math question for you. You said you had Two 36v 20Ah packs, but you also said you were using them as a 72v 40Ah?

Click to expand...

I know what you mean and all i was saying is that my speed controller is 72v 40A.

My two 36v 20ah batteries in series made a 72v 20AH battery.. Thanks for clarifying, though I do understand the distinction between amps as a rate of discharge and amphours as a capacity of the battery.

Thanks for the additional info on the ligo packs & suppliers as well.. will keep reading and keep looking!

amberwolf said:

...

Click to expand...

Another great response.. Thanks again.

& thanks to everyone else as well..

I will look at a new set of batteries to keep my ebike rolling.. Though, the truth is, since I got it, I ride my `89 KLR650 a lot more often than the ebike..

Its too bad batteries are still so expensive.. at the $2200 mark, i can buy 2 or 3 KLR motorcycles, or even a used BMW motorcycle & over 7 years, my maintenance fees will be much lower than replacement lithium batteries. ‾\_(ツ)_/‾


Thanks again!