Can I recover drained lipo?

[gwhy!1]

What I am thinking is if you can charge them to full capacity with very little current ( so no expansion, heat,explosion ) then discharge them at a very high discharge ( if a 20c then discharge at 25c ), if the lipos are beyond saving then they will get hot,expand,and pop so Im thinking this cuts out the middle man and will give a very good indication whether the lipos are fit for the bin

 

[izeman]

one thing that I have always wondered is do lipo get warm/hot as they expand just before they blow up ? .. or do they just start expanding without rising temps, I know undamaged cells have very little ( if any )temp rise when charging. Im thinking along the lines of how I used to fast charge nicads,, (many years ago) and used to terminate the charge cycle when a certain temp was reached.. i guess lipo do/can get warm because all my lipo chargers has a temp sensor input that can be connected to the battery.

yes. bad lipo get's very hot and soft. during charge and during discharge. bad lipo has a higher internal resistance. this causes the cell to become hot when you put some current into it. and heat is strictly to avoid as this (and other chemical processes) causes (the name says it all) thermal runaway.

so can a lipo be charged without fear of popping if the current is kept low enough not to get the lipo warm even if its damaged.. basically is it just the thermal runaway that causes the pressure to build up or is there something else going on as well .

that's a good question that i don't want to answer really. imho a cell be can be damaged on a chemical base and will not really take a charge. so if you charge it very slowly it will not take charge. if you charge it fast it will get hot. anyway it's dead. and if it's discharged to 0v and mostly will NOT take any charge at all. but there be pros that will argue differently and can surely add some knowledge here.

 

[e-beach]

From my experimentation, used and overly discharged cells react on an individual basis, at least with computer lico. Lico is what I am playing with at the moment.

I bought 49 used dell computer batteries and ended up with about 325 18650 cells. I have not completed the testing of all the cells yet, and I will post my results on ES later this year when I am done, but every cell reacts a bit differently from the next.

All originating voltages were as they came out of the battery packs..
In general, the ones that arrived at zero volts never charged much. Most of them didn't charge at all.
The ones from .01 to 1 volt didn't take a full charge but may have some useful purpose in a low discharge setting like a small flashlight.
Most of the ones from 1-2 volts also didn't really charge to full capacity but some did charge to 4.2v and have as much as 60% of rated capacity at .45 C discharge to 3v.

The ones from 2 volts and up did or didn't take a full 4.2v charge depending on the condition of the cell.

Some were able to charge to 4.2v and discharge at 1.9xx amp hours at .45 C until drained to 3v without getting hot. Others would not take a full charge and offered very little in capacity, like a 4.16v charge and 0.05XX amps at .45 C to 3v also without getting hot. There is no way to tell until they are charged as fully as the cell will take and then discharge them through a watt meter or balance charger with a discharge capability and watt meter.

You will never know until you try them...So charge those lipos and discharge them to see what they got in them....just be safe!

 

[dnmun]

about this chemical change stuff. i have a hard time understanding how discharging the cell to zero is gonna change the chemicals inside unless you open the cell and pour out the electrolyte. but even then the cathodic material will still be locked into the polymer matrix.

maybe someone knows how the chemicals are changed and can inform.

i thought the damage was caused by the collapse of the polymer matrix holding the cobalt oxide as the stored charge dropped below 50% down to zero. my understanding is that it shrinks in volume as the charge is depleted below that minimal level and that level i though was around 50% charge or about 3.5-3.6V resting voltage.

i have a hard time understanding how allowing the voltage to drop to zero will cause it to ignite too.

from reading real research papers, i am under the impression that the thermal runaway is promoted as the cell is overcharged and the SEI crud breaks off the surface of the anode and exposes fresh anode carbon underneath which then reacts exothermically with the electrolyte at the high potential that the electrolyte is at then, and that this exothermic reaction creates a lotta heat that causes more of the SEI crud to break off exposing more anode, and this continues rapidly until the reaction is slowed and stopped by a drop in the voltage or the current being pushed into the anode by the charger or is cooled to the point it will stop the reaction that is exposing more and more of the anode carbon surface and the charging current is terminated.

 

[dogman dan]

Take RC lico to 0v at a 5c rate, and it will puff like hell and get too hot to hold with a bare hand.

Mine didn't blow, but I sure didn't try to charge that pack back up. A slow discharge to 0v that never puffed may still take a charge and be usable. But I would not be charging it inside.

Face the music, you wrecked them. They might be usable, but they are not like new. Don't trust them a millimeter.

Lyens had a lico pack do the same thing, he got it charged, and while riding it burst into fire. That's ONE I know is in the ES record.

 

[cal3thousand]

Some of the cells on my wife's pack got down to about 2 V after being left on the bike for a few days (very slow drain). I took the time to charge them up individually and slowly and check cell IRs.

Since they were still quite close together in terms of IR, I reassembled the pack and kept and eye on it (Methods kit on it). It's been nearly 1,000 miles and I've only balanced the pack 2 times and they didn't really need it.

That's why I believe that hard abuse is more damaging than slow, soft abuse.

 

[wesnewell]

They have to have an initial charge during manufacturing. IIRC it's done over a 24-48 hour period at a very slow controlled charge. If they were slow drained and didn't puff, then I would certainly try and recover them if I didn't have the money for a new pack. Well, being the cheap sob I am, I'd try and recover them no matter what.

 

[dnmun]

when you discharge a lithium ion storage battery below the level where there is a source of free ions to transport the current then the battery becomes the equivalent of a resistor and when you push current through a resistor it heats up and that is why the lipo does the same. this happens to the lifepo4 packs too.

this is the reason that the BMS has a LVC to turn off the discharge below the point where there are lithium ions free to transport the charge through the battery cell.

the LVC is not there to limit how far you can ride, it is there to protect the battery from over discharging below the level where there are lithium ions available to transport the charge.

i realize some people do not see any need to do experiments to determine how to evaluate this since they feel the knowledge is already present in some form on the wiki or other places where the knowledge is kept.

but this is the only opportunity i have seen where it might have been possible for those of us without the special knowledge to learn how to characterize the damage done from over discharging and how to determine from measurable data if poor behavior of a battery can be traced back to it being over discharged in the past.

i have not seen this info here before but would be glad to look at it if someone can direct me to where it was done.

it does not look like the info is gonna come out of this thread either but this is now the 6th case i have read of where someone has destroyed their battery by allowing the controller to drain it to zero because there was no BMS used to cut off the discharge below LVC. this alone should be educational but will never be mentioned in all the threads on lipo.

 

[friendly1uk]

Many people have spoke about there packs early demise due to over discharging. Some that did mess about trying to recover them, and in one case he did the same pack twice.

I try to forget things that don't concern me, but iirc something crystaline forms and can link up plates(if you would call them plates). The constant (and generally ignored) expansion and contraction of the cells in use makes such hard deposits a real issue. Wearing bits and bridging gaps. It is as others have said, a fire bomb waiting to happen. They don't recover.

I realise this is an RC pack not an EV one, but what sort of controller pulls a battery down this low anyway? I mean... I made an ev pack so this can't happen to me anyway, but still, how did it happen? My bike has two separate lots of protection against this and it's all bargain basement kit. Nobody has spent less surely (About $150 on batteries and charging)

 

[wesnewell]

One has to be really negligent to drain a battery pack on a bike. Would be the equivalent of leaving your car running for days unattended. You have to turn your ignition off in your car. What's so difficult about doing the same on an ebike? And don't be fooled, if you leave the bms on your pack long enough it will also drain the battery. Just takes longer. There's many post on this forum of that happening when left for longer periods of time without charging. If anyone tells you it won't, they just don't know what they're talking about. You need to disconnect the power from the battery to the BMS before storing it for long periods of time

 

[cal3thousand]

One has to be really negligent to drain a battery pack on a bike. Would be the equivalent of leaving your car running for days unattended. You have to turn your ignition off in your car. What's so difficult about doing the same on an ebike? And don't be fooled, if you leave the bms on your pack long enough it will also drain the battery. Just takes longer. There's many post on this forum of that happening when left for longer periods of time without charging. If anyone tells you it won't, they just don't know what they're talking about. You need to disconnect the power from the battery to the BMS before storing it for long periods of time

Let me save him some time:

Dennis puts a switch on ALL of his BMS' :wink:

 

[wesnewell]

And that switch will do you no good if you forget to turn it off too.

 

[Skedgy Sky]

When I started with ebikes I initially ran 4 turnigy 6s 5ah packs. First ride I accidentally overdischarged all them, not to 0v but maybe 2v. I remember a cell fell to 0v next morning in one of the 4 packs. I don't think charging worked so I replaced it.
Within a couple months I was up to 12 lipo packs with the original 4 I abused.

A year and a little over 1000 miles of hard riding later I still use it but I am getting puffy cells, most likely from the original 4 packs that took most abuse when I was still learning about lipo. Depuffed two cells in a pack 2 months ago but now see more puffy cells in same pack.

I still use all my lipo from my battery, good or bad, but am paranoid of them bursting. I noticed some cells charge faster than others. ... so I say you may be able to recover them, probably not from 0v, and if you do they'll Still fail sooner than non-abused cells such as what I'm seeing after only 1000 miles of riding. Also, you'd have to be extra careful with these damaged cells afterwards.

 

[kudos]

This has been a really interesting bunch of answers.

Lots of views from 'it's now a bomb' to 'charge it up slow it'll be fine'.

Well so far I've charged and balanced the first 10S battery and it seems fine. I'm going to leave it for a few days to see if any of the cells drop V.

I've almost charged the second 10S battery too, this one seems to be in better balance than the first and also was only down to approx 2.5V per cell.

Once both are charged and balanced I'll check the IR of the cells, I'm not really sure what I'm looking for here other than hopefully they will be around the same value?

Kudos

 

[izeman]

once they all are balance charged you can let them sit for 2 or 3 days and check cell voltage several time and see how it changes. if one cell drops significantly it for sure is totally wrecked.

as most here already stated: the battery is more or less dead anyway in a sense that it will never perform even close to a good one, and it has the risk of going up in flames. but if you take the (not so deep discharge) as a range extender pack and ride it with reduced load it may serve you well.
if you want to do all of us a favor: do some more testing with it and post it. and to repeat it again: do it outside with battery in a stone or fire proof pot and a long extension cable for you balance charger.

first test: see above.
next test: discharge with some high load and see what happens. log cell voltage as you do, check temps and take videos. discharge to 3.0v and see how it goes.
last test: if it doesn't go up in flames, recharge to 4.2v again and discharge to 0v with high amp load. post results

 

[Ykick]

Once both are charged and balanced I'll check the IR of the cells, I'm not really sure what I'm looking for here other than hopefully they will be around the same value

For me good cells measure between 5-10mOhm. 10-15mOhm/cell is time for "concern" and when it reaches around 20mOhm/cell, retire from high current use.

Dunno if this is a sticky but it should be:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=48739&hilit=+IR+testing

 

[dnmun]

This has been a really interesting bunch of answers.

Lots of views from 'it's now a bomb' to 'charge it up slow it'll be fine'.

Well so far I've charged and balanced the first 10S battery and it seems fine. I'm going to leave it for a few days to see if any of the cells drop V.

I've almost charged the second 10S battery too, this one seems to be in better balance than the first and also was only down to approx 2.5V per cell.

Once both are charged and balanced I'll check the IR of the cells, I'm not really sure what I'm looking for here other than hopefully they will be around the same value?

Kudos

yes, similar internal resistance is useful to know and if there is some way to actually identify the cells that dropped the lowest in voltage when discharged, having good data here could be useful.

i think measuring the storage capacity would be a more useful determination than internal resistance. they should all have the original 5,000mAh and the spread should be less than 2% across all the cells in all the packs, just like it is when new.

if you have cells with only 4,500 mAh and some with 5,000mAh then that is really important to know because that is where the risk is.

but the risk should be minimal if the BMS is functional and can stop the charging if the cell with low capacity goes to the HVC in front of the others. just make sure the charging mosfet is still intact and the body diode is present.

the valuable thing about using the BMS and bulk charging is that it allows you to see if one of the cells is increasing in voltage much faster than the others as it charges up. any cell that shows a tendency to charge at a different rate from the others should be monitored imo. but it may charge up faster because the pack is not really balanced too so experimental procedure and some rigor in data collection will help make sure what you learn is repeatable.

the balancing charger will mask the difference in charging rates because of the difference in the design of the BMS inside the balancing charger. the balancing charger is always adjusting the balancing transistors that are on by comparison of voltages of adjacent cells by using a series of comparators that look at and compare voltages of adjacent cells and groups of cells. this is different from a normal pcm/BMS with a fixed balancing voltage setpoint.

 

[fechter]

My experience is if a cell goes below the normal cutoff voltage, the IR goes way up. If you charge them at very low current until they get back in the healthy range, the IR drops and the cells can be charged at the normal rate. If the voltage fails to come up during low current charge, the cell is probably toast.

Discharging a pack to the point where a cell gets reversed generally causes permanent failure. This is true for almost all battery chemistries.

Charging or discharging at high current when the cell is below the safe level can result in fire or puffing.
Cell puffing can be spontaneous due to defects in the cell construction as well. I've seen a couple do this.

If cells are discharged below the safe cutoff voltage but not reversed, and at a low current, they can usually be revived with slow charging.

Cells that lose voltage (self-discharge) on standing are suspect.

 

[Matt Gruber]

i've found the self discharge has stopped after a few shallow cycles on konions that were 0 for months. so don't give up too fast if they stay cool and work well.
good to charge on ceramic tile, like the bathroom floor, where nothing is flammable. i put a torch on a tile, and kept my finger on the other side! Think space shuttle tiles! ceramic deflects heat.

 

[Kingfish]

To the OP: I have made the mistake twice, and like others – can feel your pain.

Both times I was able to recover some (about 50%) fully and have continued to use them. The rest have one or more unrecoverable cells; they may take a charge but do not hold it for any length of time – and those I moved immediately to the recycle pile and tombstone them.

When bringing them back, as previously suggested I use a very low amp supply and trickle them back in increments, checking frequently for balance and discharge and heat. A slow charge does not bring about significant heat – even if a cell is dead-shorted. I have never had one bloat due to charging; only discharging hard passed LVC – which occurred frequently my first year into the hobby.

That said, even with the units I was able to recover, I know the lifetime has been reduced by some measure. Although in 4 years of cycling batteries - I have yet to experience that limit sans the ones that did not recover from my mistake.

Take care. KF

 

[dnmun]

yep, it is like a light bulb when they go through the LVC. you can be pushing amps through the pouch like crazy right down through 3, 2.5, then when you hit 2V the pouch gets hot fast. that is where all the ping pouches puff imo.

i have very briefly over discharge lifepo4 down to 1.6V or so under .4C load and charged them back up right away, but the resting voltage climbed up over 2.1V before i got the charger on it so i don't think it was damaged. but i think for lipo the damage is caused by the collapse of the structure holding the cobalt oxide, or that is what i read. that is the damage everyone talks about here that comes from discharging below 3.5V where the balance goes way off fast.

 

[kudos]

Thought I would give a little update.

I nursed them back at low amps up to full charge, then balanced.

Left overnight and there is one cell discharging out of the 20, but only a little.

Balanced again and went for a blast, seems to be as strong as ever.

The batts don't show any puffing either by the way.

I've now cycled them several times raising the charge amps each time until now I am back to charging them I always did which is approx 7.5Amps over the 20 cells, I charge with a 10S charger in parallel.

I did take all the internal resistances but lost the bit of paper but from memory they were ranging from 8 - 16.

I'm tempted to keep the pack that stays in balance, there doesn't seem to be anything wrong with it......

Kudos