is it plausible to re use "defective" E bike batteries? how would you do it?

[oldspice]

hi i have access to E bike batteries that will otherwise be sent to be disposed of. I am not 100% sure if they have just 1 bad cell, or maybe a bad BMS or what. i have only watched lots of videos of making batteries and never done it myself, but it looks fun. I am curious if you had this opportunity how you would go about seeing if these batteries can be salvaged- I know some of the entire packs for these bikes cost thousands of dollars new. should i put them on a charger as is? or dissamble the casing and test the cells 1 by one? or remove the connections and charge 1 by one? im really not sure where to begin but was hoping one of you guys could give me some good advice. thanks.

 

[OffGridInTheCity]

There are many people with interest (and time on their hands) that have used 'second hand cells' to build batteries and powerwalls. The basic idea is that many or even most of the cells in an e-bike battery (or medical battery or modem battery or laptop battery or RING pack or xxx) will still be useful even though the BMS has gone bad or a few cells have gone bad triggering BMS shut-off. I'm assuming these ebike batteries are 18650 NMC chemistry in general the process outline below. However, the concepts apply to any lithium-ion battery - e.g. test and assemble.

The basic process is....
1) Break the ebike battery apart into it's individual cells
2) Do a charge/discharge/charge test to see the capacity of each cell. There are testers such as an OPUS - https://www.amazon.com/BT-C2000-cha...y-Analyzer-Portable/dp/B00JL3XL2G/ref=sr_1_1- but it's just a process of doing full-charge and full-discharge to measure capacity however you do it.
3) If the cells have 90% or better their original capacity - then keep/sort them into 100mah capacity groups at full charge and let them sit for a few weeks. I use 95% of original capacity for my personal criteria but cells with 85% or even 80% original capacity are quite useful.
4) Let the fully charged cells sit a few weeks. If the voltage is steady (>4.10 for NMC) after 3 weeks then keep. If the voltage drops below 4.10v after 3 weeks then discard as they're self-dischargers.
5) Measure IR to make sure they're in normal range.
6) Then...... assemble the individual cells into battery(s) - solder or spot-weld into packs, hook the packs in series, and add a BMS.

Search youtube "Build 18650 battery" for an idea. For 18650 cells, it takes ~700 cells (e.g. 14s50p @ avg 2600mah/cell @ 48v) to get a 6.25kwh battery. It's labor intensive to get 700 good cells but it can be cost effective if you get a source of mostly good cells and have spare time for this kind of hobby/work.

 

[oldspice]

There are many people with interest (and time on their hands) that have used 'second hand cells' to build batteries and powerwalls. The basic idea is that many or even most of the cells in an e-bike battery (or medical battery or modem battery or laptop battery or RING pack or xxx) will still be useful even though the BMS has gone bad or a few cells have gone bad triggering BMS shut-off. I'm assuming these ebike batteries are 18650 NMC chemistry in general the process outline below. However, the concepts apply to any lithium-ion battery - e.g. test and assemble.

The basic process is....
1) Break the ebike battery apart into it's individual cells
2) Do a charge/discharge/charge test to see the capacity of each cell. There are testers such as an OPUS - https://www.amazon.com/BT-C2000-cha...y-Analyzer-Portable/dp/B00JL3XL2G/ref=sr_1_1- but it's just a process of doing full-charge and full-discharge to measure capacity however you do it.
3) If the cells have 90% or better their original capacity - then keep/sort them into 100mah capacity groups at full charge and let them sit for a few weeks. I use 95% of original capacity for my personal criteria but cells with 85% or even 80% original capacity are quite useful.
4) Let the fully charged cells sit a few weeks. If the voltage is steady (>4.10 for NMC) after 3 weeks then keep. If the voltage drops below 4.10v after 3 weeks then discard as they're self-dischargers.
5) Measure IR to make sure they're in normal range.
6) Then...... assemble the individual cells into battery(s) - solder or spot-weld into packs, hook the packs in series, and add a BMS.

Search youtube "Build 18650 battery" for an idea. For 18650 cells, it takes ~700 cells (e.g. 14s50p @ avg 2600mah/cell @ 48v) to get a 6.25kwh battery. It's labor intensive to get 700 good cells but it can be cost effective if you get a source of mostly good cells and have spare time for this kind of hobby/work.

wow thank you for the detailed reply... even added a link- super helpful.

I dont know or remember what IR stands for- whats that?

I watched videos on building batteries and it seems relatively easy- i just need to figure out where to get some of those hand held spot welders without having to make my own.


my other thought is if i can repair the batteries and put them back into original form of the bike battery, i could sell them and buy way more cells :D

 

[sunshine_eggo]

A few things to add:

IR = Internal Resistance.

1) if you find a bad cell in parallel with other cells, it's likely taken all the other cells out UNLESS any single cell fuse or fuse-able link is blown thus isolating it from the rest of the parallel group.

2) If you find cells that have discharged below 3.0V/cell, segregate them. If they've discharged below 2.5V, shit can them. For the segregated cells, it's important to assess their test results compared to the other presumed good cells. If you see consistent differences between cells > 3.0V and cells < 3.0V, likely not a good idea to use the <3.0V cells.

 

[robby]

A few things to add:

IR = Internal Resistance.

1) if you find a bad cell in parallel with other cells, it's likely taken all the other cells out UNLESS any single cell fuse or fuse-able link is blown thus isolating it from the rest of the parallel group.

2) If you find cells that have discharged below 3.0V/cell, segregate them. If they've discharged below 2.5V, shit can them. For the segregated cells, it's important to assess their test results compared to the other presumed good cells. If you see consistent differences between cells > 3.0V and cells < 3.0V, likely not a good idea to use the <3.0V cells.

100% Agree with this.

 

[oldspice]

thanks for the replies yall. I have since watched a buncha repair videos on youtube where some ppl are experiencing only bad bms systems that only need to have the reset button pushed! Ha that would be exciting...

 

[oldspice]

2) Do a charge/discharge/charge test to see the capacity of each cell. There are testers such as an OPUS - https://www.amazon.com/BT-C2000-cha...y-Analyzer-Portable/dp/B00JL3XL2G/ref=sr_1_1- but it's just a process of doing full-charge and full-discharge to measure capacity however you do it.

i just noticed this item you linked is for nimh / etc and not lipo types- i think most e bike batteries are lipo- i assume thats a problem unless you mean the discharge function can work on any?

 

[sunshine_eggo]

i just noticed this item you linked is for nimh / etc and not lipo types- i think most e bike batteries are lipo- i assume thats a problem unless you mean the discharge function can work on any?

Most e-bikes are 18650 NCA. "LiPo" is generic for 3.7V chemistry including NCA, NCM, LMO, LiPo, etc.

I don't see that unit indicating it's good for 18650.

I have this one:

Amazon.com

Works great for 3.7V 18650. I use it for those and Eneloop NiMH AA/AAA recharging.

 

[KITROBASKIN]

If you have an ebike that you want additional packs, just be careful. As far as selling/giving those kinds of packs to other people? Potential venting with flame could result in tragedy you do not want to play a part.

Spot welders for 18650 battery connections are not that expensive, but there is this for experimentation and possible small scale use:

VRUZEND DIY Battery Kit

vruzend.com

Note that some kind of compression will help connectivity with this system but there are limits, and a mobile application invites issues with vibration. 18650 diameter needs to be confirmed.

You have an opportunity to have fun and learn. Distractions while pack building are not good.

 

[OffGridInTheCity]

i just noticed this item you linked is for nimh / etc and not lipo types- i think most e bike batteries are lipo- i assume thats a problem unless you mean the discharge function can work on any?

Apologize - I ment the OPUS BT-C3100 style like this - https://www.ebay.com/itm/354172501981 It can do 18650 NMC (3.0v to 4.2v) but also has an internal switch you can do LifePO4 cylinder cells as well and was a very commonly used tester a few years ago. But there are others.

 

[sunshine_eggo]

Apologize - I ment the OPUS BT-C3100 style like this - https://www.ebay.com/itm/354172501981 It can do 18650 NMC (3.0v to 4.2v) but also has an internal switch you can do LifePO4 cylinder cells as well and was a very commonly used tester a few years ago. But there are others.

For the same price, I'd choose the MiBoxer I linked. It reports IR as well, and it handles 3.2/3.7V chemistry automatically... plus it does 8 at once!

 

[oldspice]

anyone know how i could test a pack to see if its good before disassembling it into individual cells? or testing those individual cells before removing the nickel? I feel like if its all connected you cant test individuals??

hoping to find salvageable packs for ebike use.

 

[OffGridInTheCity]

anyone know how i could test a pack to see if its good before disassembling it into individual cells?

You can test a pack as a whole - you need the ability to charge and discharge+measure. You can buy a charger compatible with the voltage range of the pack for the charge side. Build a DIY load/with shunt+meter to measure ah as you discharge. Or buy a combo charge/discharge tester.

or testing those individual cells before removing the nickel? I feel like if its all connected you cant test individuals??

For cells in parallel, you can test the group and divide the results by the number of cells. However, if you have a less than ideal cell in the group it won't be visible. So you wouldn't catch slightly self-discharging or hi IR (internal resistance) cells so easily.

If cells are in series, you could test individual cells (or groups of parallel cells) while still connected in series by attaching directly to the + and -. But you don't want the overall battery active in terms of charge/discharge while doing this and you need to bring the test cell back in balance with the other's in series before using the overall battery.

For longevity of the future battery - it's best to test the individual cells. However, not all batteries need maximum longevity for the intended use - so it's a judgement call.

@jehugarcia youtube channel has long demonstrated hooking up 20 or 40 or 60 (or more) ebike or other packs in parallel 'as is' to make a powerwall - and what does it matter if a few are subpar as long as the overall contraption works/meets-your-needs. It's a perfectly rational approach to save a BUNCH of time and if the BMSs are working on each pack, then I don't see a big risk.

 

[oldspice]

For the same price, I'd choose the MiBoxer I linked. It reports IR as well, and it handles 3.2/3.7V chemistry automatically... plus it does 8 at once!

how do i view the IR on that unit? it appears to not have testing functions?

 

[oldspice]

does anyone know how to un-weld the nickel off the 18500 batteries? i was pulling it off with a pliers and the metal touched the edge of the battery and created a spark... i didnt think + and - were only a few millimeters apart on these batteries.

 

[sunshine_eggo]

how do i view the IR on that unit? it appears to not have testing functions?

When you charge, it reports an IR value that alternates with the reported charge current. I have some high resistance 18650 for a headlamp currently in it. They're measuring 245 and 225mΩ., respectively. When I swap their positions in the charger, they read 244 and 222. It's not always that consistent, but it's typically within ±5%.

 

[sunshine_eggo]

does anyone know how to un-weld the nickel off the 18500 batteries? i was pulling it off with a pliers and the metal touched the edge of the battery and created a spark... i didnt think + and - were only a few millimeters apart on these batteries.

The entire outer casing is the (-). Gotta be very careful when doing the (+) end.

I just carefully use pliers-like nail clippers to nip/peel off the welded strips as well as the shards that may remain.

 

[oldspice]

When you charge, it reports an IR value that alternates with the reported charge current. I have some high resistance 18650 for a headlamp currently in it. They're measuring 245 and 225mΩ., respectively. When I swap their positions in the charger, they read 244 and 222. It's not always that consistent, but it's typically within ±5%.

if its at 100% charge does it stop reporting it? i took your advice and got one, just need to get used to its functions i guess. Thanks for the replies!

EDIT: it appears it only shows while charing- its showing 1 amp / 60-65 (is that bad ir?) do you want higher or lower number to show a good battery/?

 

[sunshine_eggo]

if its at 100% charge does it stop reporting it? i took your advice and got one, just need to get used to its functions i guess. Thanks for the replies!

EDIT: it appears it only shows while charing- its showing 1 amp / 60-65 (is that bad ir?) do you want higher or lower number to show a good battery/?

There's a short delay as the cell is inserted. As it's charging, it will alternate between displaying the charge current and the IR. When charging is complete, it will alternate between "Full" and "XXXmΩ"

I wait until they're fully charged before recording IR. It usually improves vs. the number reported when charging begins.

 

[OffGridInTheCity]

The entire outer casing is the (-). Gotta be very careful when doing the (+) end.

I just carefully use pliers-like nail clippers to nip/peel off the welded strips as well as the shards that may remain.

You'll get better. I also use small wire cutters to wiggle, cut, leverage off the metal from the end of the cells. Since I solder, it doesn't have to be perfect.

An 18650 can is all negative except the + end. At the + end, it's the ~1/16" outer rim that's part of the negative can. So avoid leveraging the tool from center to the edge as it will scrape thru the protective layers. Pull more than pry when doing + end.


I also jump when sparks fly but the cell won't explode - it's just a bit hard on it and of course a prolonged short will create heat pretty fast.