Lifepo4 fully discharged... can it be revived?

Joined
Jun 22, 2015
Messages
6
Hi All,

Apologies up front for a basic question....

I have Lifepo4 48V EBay special that powers my home made ebike.

6 months ago I stopped using the bike and now I went to charge it at its nominal 60V and it is not retaining its charge at all.
I read somewhere that if the cells fall below 2.6v then it's not wise to try and recover the battery cells.

The battery has been connected to the bike for all this time draining. Yes I know this is a stupid rookie mistake but I guess I am wondering if there is a recommended fix. There is a built in BMS and I am looking for guidance before I start to break into the blue shrink wrap.

Any advice would be appreciated even if it is 'go but a new one'.

Thanks in advance Chris.
 
This particular chemistry in general is "less likely" to burst into flame because of the problem you've got, but none are immune to it, so be aware of that first.

Next, it is possible (likely) that whichever cells went too low will have either (or both) lower capacity than before, and less current delivery capability (more voltage sag under load), compared to how they were before.


If you're willing to accept that, and still want to try reviving it, you'll need to open it up far enough to reach the BMS and it's wires that run to each cell. At the connector on the BMS you can measure with a voltmeter from the main negative output to the first pin on that multiwire connector, then from that first pin to the next one, and so on until you get to the last pin, and measure from that to the main positive output wire.

Write down all those voltages in that order, and you'll have a list of the cells in the pack.

Probably it's just whichever cells are used to power the BMS (usualy the first few), so you can then manually charge each of those, slowly, until they match the other cells.

This can be done with a "lab PSU", which has both adjsutable voltage, and adjustable current limit. This is an easy way to be able to charge single cells of any chemistry, or even the whole pack if you get a PSU that has a maximum voltage high enough to do that. It can also be used for troubleshooting other electrical things, including ebike stuff like hall sensors or throttles, etc.

Or you can use a single-cell LiFePO4 charger, but most of the ones I ahve seen arent' adjustable, so those will simply fully charge a single cell as fast as they can. If you want to be gentle to the cells that are too low, this probably wont' do that.


There's other ways, too, if you look around at posts that mention single cell charging or charger.



Once youve got all the cells about the same voltage, and above whatever the BMS cutoff is, you can then just run the regular charger to fill up the pack, and then test it to see if it still does what you need it to do.

If it doesn't, you could replace the cells that were run too low, and probably fix it, or you can replace the whole pack.
 
Hi Amberwolf,

Thanks for the reply. I don't have a lab PSU but I know what you mean.

I appreciate the info, I will attempt a revival based on your post.
 
If you don't have a lab PSU, but have a computer PSU, there's three voltage "rails" you can draw off inside - 12, 5v, and 3.3v. The 12v rail is always yellow, the 5v rail is always red, and the 3.3v rail is always orange. Doesn't matter which ground (black) you use, they're all common.

You can use the 3.3v rail to bring each individual cell up to the "right" voltage. The 3.3v rail typically has a lot of current behind it, so you can quickly charge even fairly large capacity cells.

If you're real brave, once the current starts dropping, you could even use the 5v rail to bring the cell voltage to 3.7v. Just make sure you don't leave that unattended. While LiFePo4 can cop some over-voltage, 5v is way too much, and you're already dealing with compromised cells.
 
Thanks Sunder. I think I have a computer PSU somewhere. Otherwise it's $75 for a bench PSU that would be handy to have around.

I am wondering if the battery pack is worth reviving as it's very heavy at 4.5kg. The bike handles poorly with that weight on the back. I realise it's always a compromise between weight and power.

I upgraded my controller to peak at 40A output. I was thinking I could get a smaller capacity cylinder type battery that mounts on the bike frame not pannier.

You mention the battery cells are already compromised so I guess I won't be achieving high current at maximum voltage?

Thanks for your input.
 
These are a great investment in EV life, under $40 on ebay. I have used NiCd setting to revive cells then switch to the balance charge LiFe.
 

Attachments

  • IMG_5025.JPG
    IMG_5025.JPG
    62.4 KB · Views: 6,950
One thing to keep in mind about any of the charging methods that don't let you reduce the current to something low (in the 100-200mA range probably) is that the (much) higher charging currents could be beyond what the compromised cells can handle, resulting in further problems (or worse).

Some of them (like teh computer PSUs) can often provide currents (far) beyond what even a normal cell is spec'd to be charged at.

As for weight and handling, the best bet is to not put that weight on teh tail, but put it in the middle of the bike (triangle, etc), or move it to a side pannier, even if it is all on one side, then if it is down low around the axle it is still easier to ride with than up high over the rear wheel.
 
Thanks again guys, all good information.
This is my first and only ebike.(so far) It's a mid drive mounted on a Giant NRS X3 full suspension bike. The battery is so big and heavy the rear rack was the only option.
I thought if I build my own battery I can design the arrangement to fit within the limited triangle space and hence lower on the frame and part of the bike.

I looked on eBay for a generic cylinder type battery, I am looking at around A$300 for a 48v 10aH.

Still not 100% sure about attempting a revival as it's cost of trickle charger / bench supply, time, compromised cells and a potential fire risk hazard vs A$300 or whatever the cost and time to build my own pack.

Thanks again for your support.
 
Get to the cells and write down first. Free to look.
1. 2.85v
2. 3.25v
3. 2.98v


16. Xxx volt
I have 4 voltpheks lifepo4 single cell that stop at 3.65v @ 2amp which is to high a charge rate if below 2.8v a cell phone charger can work. But even thou printed on side says 5v 200mah TEST as it could be 7volt. Seat and watch if going to the bathroom unplug. Only seat and watch.
When you look for a small battery for a 40amp controller you most buy name brand and the quality ones. Samsung 25r or ? Get ones over rated for your needs. Lots of crap out there. Ask first.
 
You don't have much to lose by taking apart the pack and see if you can measure the individual cell voltages. If any cell is too low, the BMS will prevent any output. You might just have one low cell that's tripping the BMS.

Even if you get a new pack, it's nice to have a spare one that works.
 
Yes nothing to lose but some danger.

New development. I attached some SLA's that I was given and found that any load over 20A throws the drive chain on my mid drive. It has done this since I upgraded my controller and the drive cog became loose. I tightened it but it still throws the chain??

It's pointless having a battery if the bike regularly throws the chain.
1st priority, fix the chain and utilise the SLA's

2nd revive the LifePo even if used as a spare.

I will keep you posted, thanks for your advice.
 
Sometimes cells can be revived but once they drop below the saftey voltage you must revive them with very low amounts of power until they reach the safety voltage. So you need a way to limit the current until this voltage is reached. A BMS helps to balance the cells and you might find one in the battery but they can also be bought seperatly online. They are also in balance chargers but most chargers don't balance above 8S perhaps 10S. Some do only 4S balancing. By buying a BMS of your choice, with a little skill you can build cells up into a battery with a voltage of your choice.
As to weather the old battery can hold it's voltage or be any good at storing power? You will just have to test it. There might be some good cells inside worth salvaging. If you have lots of time you can build up a collection of similar good cells and use them in future projects. Lead Acid are heavy and don't suit that many E-Bike enthusiasts. But i know a few people who use them, ok in trikes and the like.
 
Charlie_Echo_Foxtrot said:
any load over 20A throws the drive chain on my mid drive. It has done this since I upgraded my controller and the drive cog became loose. I tightened it but it still throws the chain??
Perhaps you should post up your build and details, so the builders with experience in such things can help you with that, too. ;)
 
If you do go back to working on the battery buy a digital volt meter from local hardware or auto parts $20 or less, valuable tool and check each bms lead for specific pack voltage. Your thread reminded me of a pack I had go bad so I worked on it on my build thread and it will be revived. as for your chain please post lots of detailed pics and give us a better idea of your situation. More details = more help.
 
I have revived a few 100 LIFEPO4 cells. Measure them at millivolt to get more accurate info. In my opionion cells that are less then 0,5V are not worth reviving, But cells above 1V can often have 70-90% of its original capacity! USE them NIMH setting on the charger and charge up to 2V then switch to LIFE.
 
I have been reading some older threads pertaining to Lifepo4 batteries as I know next to nothing about them. I have a 36v 8ah battery in one of those Silverfish cans, the thing only wants to take a charge to about 38.7v, the pack is pretty old, 5+ yrs old, motor is supposedly 260 watt geared hubmotor is shutting down when riding up hills so I'll guess the pack can't put out enough amps any more to ride anything but flat ground. Rode it across town and back this afternoon and it shut down about mile 6-7 or so, even pedaling the whole time on flat ground. This is on a folding bike and due to it being a geared hub motor it rides real good with the motor off so getting back home was no problem, only walk of shame was a couple hills right below home base. Battery showed 36.1v when I got home.

I opened the case up in an attempt to pull things out, inspect things and maybe check the cell voltages, everything is packed in there beyond tightly, foam rubbery stuff holding the cells in really firmly so it's going to take some force to dislodge things. Curious if I should try to rebuild the pack using the BMS with new Lifepo4 cells adding some more capacity, what would my options be for doing that?

This isn't a performance bike, just want to be able to cruise with a light bit of assist and some help up the hills, the bike is my Clark Kent mild mannered normal ebiker bike, kind of fun to get on the bike path and ride alongside the Pedego People once in a while, nobody even notices me.
 
You need to know the state of charge of your battery. A 36 volt battery lifepo4 has 12 groups of cells . You start by knowing the voltage of each cell group 1 through 12. You need to write them down on a piece of paper this is a must like this.
1. 3.33v
2. 3.40v
3. 2.98v

12. Xxx volt
This can be checked at the sense wires going to the BMS starting at the negative end as number one. No need to take battery out of the case at this time. You do need a multimeter. Free at Harbor Freight with coupon if you buy zip ties for $2 or razor blades $1.
 
999zip999 said:
You need to know the state of charge of your battery. A 36 volt battery lifepo4 has 12 groups of cells . You start by knowing the voltage of each cell group 1 through 12. You need to write them down on a piece of paper this is a must like this.
1. 3.33v
2. 3.40v
3. 2.98v

12. Xxx volt
This can be checked at the sense wires going to the BMS starting at the negative end as number one. No need to take battery out of the case at this time. You do need a multimeter. Free at Harbor Freight with coupon if you buy zip ties for $2 or razor blades $1.

Yes testing the individual cell voltages was my goal when I opened the case, I'll try to post a photo tomorrow of what it looks like, Seems like they did a pretty fair job of trying to prevent trying to repair it and I didn't want to get too rough trying to pop it open to access the wires to the BMS without seeking some guidance.
 
Reading after being on charger all night 38.9v
Hal44~iBlx.JPG

Opened up case by removing screws and pulling plastic ends off, top end
63Ljt5uqrk.JPG

Bottom end
QfeuJlDCDb.JPG

Pulled insulating sheet back on bottom
MI_wpDO47z.JPG
 
Pulling the insulating sheet back on the top end I see 3 screws holding what appears to be a BMS, will be loosening the screws next.
aaiqvxrh77.JPG
 
edrift.eu said:
I have revived a few 100 LIFEPO4 cells. Measure them at millivolt to get more accurate info. In my opionion cells that are less then 0,5V are not worth reviving, But cells above 1V can often have 70-90% of its original capacity! USE them NIMH setting on the charger and charge up to 2V then switch to LIFE.

A123 says in one technical doc that 0.5V is the absolute minimum voltage for their LiFePO4.

So, what you say seems to be good.

Thierry
 
After screws were loosened I was able to pull the BMS forward disconnecting the pins. I count 12 pins on the plug.
ES6rN&Dq!2.JPG
 
Back
Top