Balancing 300 18650 Batteries :/

Joined
Jun 26, 2021
Messages
10
Hey Guys,

i am fairly new to the whole spiel and I got (via a friend) 300 18650 Cells, all around 1600-1750 mAh.

Now, before building a battery-pack i think it would be best to balance the cells.
Does anybody has a viable idea to balance these cells without putting all of them into a charger for ever ?

Best

SackofFluor!
 
Step 1. You would have to check for voltage and sort by voltage 3.35v or 4.1v or 3.0v you can write on the the cells with a pen. But at 300 you would need different boxes. remember those cells have a lot of stored energy. Those below 2.5 v or better 2.8v after that recycle the cells.
 
Yes sorting to create even cell-groups should be based on capacity not voltage

faster self-discharge a strong negative

ideally also resistance

 
well I R is also good in indicator but how many hours do you want him to spend on 1,500 mah cells
Yes voltage capacity and internal resistance
how big of a battery are you thinking of building and what's it for ?
 
Thank you for your replies so far....

So i got them from a company that couldnt/didnt sell them. they are all individually measured and range from 1600-1799mAH
So they are kinda a sorted.

The idea with the tinfoil is brilliant! i do assume that you do the same at the other side as well :D

In general, this should be my first start for a larger powerwall that I want to charge over comparable long times using solar, so I can use the stored energy on days where i have a high load because of my beer brewing.

So far I was aiming for a 7s 40p setup, but I think that would not be enough, since the batteries would then be discharged with 2A and i think that is a little bit too much.

therefore i think i will end up with something like 7s 100p or maybe even 140p.

In general, it's my first venture in this topic, so if there are problems or some cells die... well it is what it is...
 
SackofFluor said:
i am fairly new to the whole spiel and I got (via a friend) 300 18650 Cells, all around 1600-1750 mAh.
If all 300 18650 cells (same brand?) have same [original] rated mAh capacity and same MCD amp rating then goatman's reply w/photos is applicable. Who told you or how did you determine these 300 salvaged/used 18650 cells presently range from 1600-1799 mAh capacity?

If all 300 cells are from 2 or more different brands having different mAh rated capacity and MCD amp rating then John61ct's reply is more applicable ...
john61ct said:
Yes sorting to create even cell-groups should be based on capacity not voltage
For example Voltaplex lists ten 18650 cell brands that range from 1600 to 1800 mAh capacity with different MCD amp ratings ... https://voltaplex.com/lithium-ion-battery/18650-battery?limit=36&s_capacity_max=1600-1800 ... There are also other off-brands within this same 1600-1800 mAh rated capacity. You first need to sort the 300 according to their brand mAh rated capacity and MCD amp rating. Who told you it's ok to use different 18650 cell brands having different mAh rated capacity (1600-1799 mAh) and possibly different MCD amp ratings in the same powerwall?
SackofFluor said:
Thank you for your replies so far....
We need more clarification about which of the above two scenarios is applicable before we can provide more specific advice.
 
eMark

I've asked before, please do not refer to my name nor my posts in yours.

Especially agreeing with me!
 
eMark said:
If all 300 18650 cells (same brand?) have same [original] rated mAh capacity and same MCD amp rating then goatman's reply w/photos is applicable. Who told you or how did you determine these 300 salvaged/used 18650 cells presently range from 1600-1799 mAh capacity?

If all 300 cells are from 2 or more different brands having different mAh rated capacity and MCD amp rating then John61ct's reply is more applicable ...
For example Voltaplex lists ten 18650 cell brands that range from 1600 to 1800 mAh capacity with different MCD amp ratings ... https://voltaplex.com/lithium-ion-battery/18650-battery?limit=36&s_capacity_max=1600-1800 ... There are also other off-brands within this same 1600-1800 mAh rated capacity. You first need to sort the 300 according to their brand mAh rated capacity and MCD amp rating. Who told you it's ok to use different 18650 cell brands having different mAh rated capacity (1600-1799 mAh) and possibly different MCD amp ratings in the same powerwall?

Neither of these concerns are material for a salvage pack. Also, this isn't a tabloid paper -- don't write like one.

Anyway, just cull overdischarged or overcharged cells per zip, balance per goatman, and then equally distribute the cells into parallel groups by whichever metric is available -- tested mAh, in this case.
 
No, there are definitly different brands!

Some of them still have their Modelnumbers:

like:
ICR18650-22P
ICR18650-22F
HW 18650 3.7G OA09
or CGR18650CG

But then there are also som which don't have any distinguishable markings, those cells have a dark brown or red color.
The company that gave them to me noted their mAH on each cell individually.

So Those cells were of higher capacity beforehand but now seem to have the measured mAhs.

Nobody "told me its ok to use different 18650 cell brands...." that's why I asked and wanted to gather more information...

Thanks guys!
 
Circumstances and preferences vary widely.

Personally I would not bother putting that much work into secondhand.

But thousands do.

At very high C-rates the pack you build may only give you a dozen or two cycles

Used for a USB power bank for weekend camping might last a couple years.
 
SackofFluor said:
In general, this should be my first start for a larger powerwall that I want to charge over comparable long times using solar, so I can use the stored energy on days where i have a high load because of my beer brewing.

So far I was aiming for a 7s 40p setup, but I think that would not be enough, since the batteries would then be discharged with 2A and i think that is a little bit too much.

therefore i think i will end up with something like 7s 100p or maybe even 140p.

In general, it's my first venture in this topic, so if there are problems or some cells die... well it is what it is...
Would be an interesting learning experiment if all 300 cells were same cell brand as in goatman's photos.
SackofFluor said:
No, there are definitly different brands!

Some of them still have their Modelnumbers:

like:
ICR18650-22P
ICR18650-22F
HW 18650 3.7G OA09
or CGR18650CG

But then there are also som which don't have any distinguishable markings, those cells have a dark brown or red color.
The company that gave them to me noted their mAH on each cell individually.

So Those cells were of higher capacity beforehand but now seem to have the measured mAhs.
Using those salvaged cells of questionable quality and different specification (e.g. 1600-1800 mAh capacity and 4.4-10 amp MCD rating) is possibly just as much or more of a safety risk as a viable solution. When cells die they don't always go silently without cause and effect of further damage.
 
+1
300 cells, you can build same cells in one pack
build multiple packs of each battery brand and then parallel them together
mixing and matching in one big pack would be a nightmare
 
SackofFluor said:
Nobody "told me its ok to use different 18650 cell brands...." that's why I asked and wanted to gather more information...
It's perfectly okay -- just try to equally distribute the different cells among the parallel strings.
 
So you would make equal splits of the different cell types and then build packs that have "equally" distributed those splits within the individual packs?
Thank you! Most likely its a very good idea to also to do most of the work outside, at least in the beginning phases :D
 
eMark said:
Using those salvaged cells of questionable quality and different specification (e.g. 1600-1800 mAh capacity and 4.4-10 amp MCD rating) is possibly just as much or more of a safety risk as a viable solution. When cells die they don't always go silently without cause and effect of further damage.
If equally distributed, different cells are not any more risk than any other salvaged cells.
 
goatman said:
+1
300 cells, you can build same cells in one pack
build multiple packs of each battery brand and then parallel them together
mixing and matching in one big pack would be a nightmare
This is incorrect. Building serial packs and then paralleling them introduces balance problems that can only be resolved by paralleling the packs at the parallel string level, at which point you could have just built a parallel-first pack to begin with.
 
SackofFluor said:
So you would make equal splits of the different cell types and then build packs that have "equally" distributed those splits within the individual packs?
Careful of the terminology here. Individual cells are paralleled into strings. Strings are seriesed into packs. Packs can be serised, but parallel packs should be paralleled at the string level.

You're only building one pack: split the different cell types equally among the parallel strings.

SackofFluor said:
Thank you! Most likely its a very good idea to also todo most of the work outside, at least in the beginning phases :D
There are millions of packs built of salvaged cells. The danger, salvaged or not, is in improper unprotected overcharge and/or overdischarge (including shorts).
 
fatty said:
goatman said:
+1
300 cells, you can build same cells in one pack
build multiple packs of each battery brand and then parallel them together
mixing and matching in one big pack would be a nightmare
This is incorrect. Building serial packs and then paralleling them introduces balance problems that can only be resolved by paralleling the packs at the parallel string level, at which point you could have just built a parallel-first pack to begin with.
Well, at least if similarly-worn cells are all segregated

the whole pack can be discarded / recycled when it reaches EoL,

leaving the healthier ones to continue using a few months more
 
The fact you're messing with scrapped cells is what increase the risks

much more than the pack architecture

discarding the lowest capacity ones will help.

But most important is accurate charge termination

ideally at 4.10Vpc or lower, maybe 4.15 but that hardly is a significant capacity increase

with a reliable and adjustable Smart BMS to give failsafe redundancy at say 4.2Vpc when the charge controller fails.

The need for early LVC is to extend longevity, not so much safety as such
 
john61ct said:
well, at least if similarly-worn cells are all segregated

the whole pack can be discarded / recycled when it reaches EoL,

leaving the healthier ones to continue using a few months more
True, but discarding a whole sub-pack is desirable in a weight/volume-constrained application, like a vehicle.

In a static installation like a powerwall, with no weight/volume penalties, there would be no reason to discard a sub-pack, since capacity declining under 80% original (or any arbitrary state of health) is still usable -- just replace individual cells that fail through self-discharge, or replace entire pack when self-discharge failure rate becomes unsustainable.
 
john61ct said:
The fact you're messing with scrapped cells is what increase the risks
much more than the pack architecture
Much more? Charge profile greatly influences this comparison. A salvage pack consistently balance charged to 4.15VPC is probably less risk than a bunch of series-first packs paralleled, which come out of balance and are then blindly series charged.

john61ct said:
discarding the lowest capacity ones will help.
Probably good preventative maintenance. I'd do this as some standard deviation from the mean for that cell type.

john61ct said:
But most important is accurate charge termination

ideally at 4.10Vpc or lower, maybe 4.15 but that hardly is a significant capacity increase

with a reliable and adjustable Smart BMS to give failsafe redundancy at say 4.2Vpc when the charge controller fails.

The need for early LVC is to extend longevity, not so much safety as such
Agreed
 
Back
Top