Homemade Battery Packs

On my way to build a 3s50p battery pack:

1) Charging: Open more than 150 18650 recycled laptop cells and charge them to 4.20V

2) Bleed down: Set for "bleed down" 4/5 months, then segregate cells determined by voltage

3) Capacity test: Discharge cells in pair with Imax B6 at 1A and write down capacity, then charging back to 4.20V

4) IR-test (Internal Resistance): Measure voltage using a multimeter before the Capacity test and after 20 seconds discharging at 1A, then you have the voltage drop: [unloaded V - loaded V = V drop]. Then use this formula to get the cells internal resistance: [V drop / Amp x 1000 = milli Ohms]


The Capacity test is a little time consuming and my Imax can`t discharge more than 1A. I hope the IR-test is proper. Most of the cells is discharged in pair but some as single cells, does that change the results? Give med some feedback please.

One example:
Cells that "bleed down" between 4.10V and 4.00V:
Best pair: 40mΩ and 4458mAh
Worst pair: 240mΩ and 1458mAh

kje said:

.......

2) Bleed down: Set for "bleed down" 4/5 months, then segregate cells determined by voltage

3) Capacity test: Discharge cells in pair with Imax B6 at 1A and write down capacity, then charging back to 4.20V

......

Click to expand...

Don't waist you time with a "bleed down test". Not all bad batteries will bleed down. Just charge the cells and discharge them through a watt meter. If you are using your B6, just know that it is built to be inaccurate, (it is built to keep the kids from over charging and over discharging their lipo) but if you use it for all your cells then a consistent record can be kept. You may notice that the cells that reach 4.2v have a better chance of being useful cells, while cells that don't take a full charge will have limited capacity and need to be properly recycled/discarded. If you also use a separate DMM in parallel with the B6 you may get more accurate readings.

kje said:

On my way to build a 3s50p battery pack:

The Capacity test is a little time consuming and my Imax can`t discharge more than 1A. I hope the IR-test is proper. Most of the cells is discharged in pair but some as single cells, does that change the results? Give med some feedback please.

Click to expand...

For direct comparison against 1000mA discharged pairs, singles should be discharged at 500mA - for capacity comparison ...
IR compared against other singles only.

e-beach said:

Don't waist you time with a "bleed down test".

Click to expand...

Self discharging cells must be removed!
This is vitally important if you want a reliable - stable battery build.
If you don't care about reliability or stability you must be concerned about safety!
Cells with noticeable self-discharge are defective, unreliable, heat producing and possibly dangerous!
Performing a simple bleed down test eliminates these lemons.

e-beach ...
who has never built a battery pack ...
seems to be promoting building packs with obviously defective cells!

I am not sure of his reasoning ... or lack thereof ... but he is adamantly, again and again and again, against performing the simplest, most basic-important step in building a safe reliable battery pack.

He also recommends discharge testing LiCo cells to 2.65V which is liable to severely damage, possibly destroy, them.

So ...
Please!
Ignore him!

Perhaps a Mod should monitor his posts and edit unsafe recommendations ... or at least PM him and try to talk some sense into him? Have him remove his unsafe recommendations ... at least until he has actually, successfully, built, tried and proven his method.

I certainly do not want someone promoting obviously unsafe building methods in this thread!

Another question from the example above:

kje said:

One example:
Cells that "bleed down" between 4.10V and 4.00V:
Best pair: 40mΩ and 4458mAh
Worst pair: 240mΩ and 1458mAh

Click to expand...

Can I use the best pairs from the segregated cells that "bleed down" between 4.10V and 4.00V? There was a lot of good capacity and low IR...

kje said:

Another question from the example above:

kje said:

One example:
Cells that "bleed down" between 4.10V and 4.00V:
Best pair: 40mΩ and 4458mAh
Worst pair: 240mΩ and 1458mAh

Click to expand...

Can I use the best pairs from the segregated cells that "bleed down" between 4.10V and 4.00V? There was a lot of good capacity and low IR...

Click to expand...

As stated earlier, I would build as a separate pack and charge to 4.0V.
Use as emergency "limp home" pack.
Or ...
Charge to 4.10V immediately before use and "add" to major pack 4.10V with quick connectors, T Plug, or other..
But I would not permanently build into a single pack.

kje said:

The number of cells after "bleed down":
4.20V -> 4.15V: 101
4.15V -> 4.10V: 8
4.10V -> 4.00V: 43
4.00V -> 3.90V: 10
3.90V -> 3.80V: 17

Click to expand...

Build Best 99, of the "4.20V -> 4.15V: 101" into your Primary Pack.
Best 51 or 48 etc. of the
"4.15V -> 4.10V: 8
4.10V -> 4.00V: 43"
into secondary pack.

You can always add more cells to either pack later on ... as acquired.

Sounds like a good strategy.

Pairs with low capacity or high IR will be separated and tested individually. I assume one cell could be bad but the other one good to use?

DrkAngel said:

e-beach said:

Don't waist you time with a "bleed down test".

Click to expand...

Self discharging cells must be removed!

Click to expand...

All cells that are past their useful life will be exposed by discharging them through a watt meter.

This is vitally important if you want a reliable - stable battery build.
If you don't care about reliability or stability you must be concerned about safety!
Cells with noticeable self-discharge are defective, unreliable, heat producing and possibly dangerous!
Performing a simple bleed down test eliminates these lemons.

Click to expand...

Yep, and as stated before, not all bad cells bleed down, so to best determine which cells are suitable for usage simply discharge them through a watt meter. All bad cells will be exposed.

e-beach ...
who has never built a battery pack ...
seems to be promoting building packs with obviously defective cells!

Click to expand...

Pinocchio your nose is gorwing!.....Because this statement is just not true!

I am not sure of his reasoning ... or lack thereof ... but he is adamantly, again and again and again, against performing the simplest, most basic-important step in building a safe reliable battery pack.

Click to expand...

Because simply charging and then discharging a cell, or cells through a watt meter exposes all cell characteristics necessary in determining which cells to use in a safe pack.

He also recommends discharge testing LiCo cells to 2.65V which is liable to severely damage, possibly destroy, them.

Click to expand...

Nope, I never recommended it, I said I will do it to my next pack as a way to test it.

So ...
Please!
Ignore him!

Perhaps a Mod should monitor his posts and edit unsafe recommendations ... or at least PM him and try to talk some sense? Have him remove his unsafe recommendations ... at least until he has actually, successfully, built, tried and proven his method.

I certainly do not want someone promoting obviously unsafe building methods in this thread!

Click to expand...

The safest way to test used LiCo is by charging them with a low current smart charger and then discharging them at less then 1/2 C through a watt meter. Something I am sure the Moderators understand regardless of what the OP understands.

Sorry ... you made it sound like you have never tested cells or built a battery pack !

e-beach said:

A question on your experience with recycled 18650 batteries...

I recently bought 49 dell laptop battery packs in 2 lots. One lot was 27 packs and the other 22 packs. The total price was just under $70 USA for all the packs.

I have yet to fully count the total number of batteries but it's about 325 used 18650's. Mostly Samsung 2.600 amp hour cells but also Sony's LG's and others I have yet to identify.

Right now, I have them separated into 3 groups.
A: Those that are above 3.25 volts,
B: Those that are 3.24 to 2 volts
C: and those that are 1.99 to zero volts.....I have about 113 of these.

At this point my question is what should I do with the 113or so that are 1.99 to 0 volts. They may be the dangerous ones and I am wondering if they are worth trying to revive with a slow charge to see if any are salvageable.

From your experience is it worth trying to salvage low voltage cells?

Thanks!

Click to expand...

Once again ... sorry!
Please tell us about your battery builds.

Yep, I asked that question....Last year :lol: ( Sun Nov 03, 2013 6:34 pm ) :lol:

DrkAngel said:

Sorry ... you made it sound like you have never tested cells or built a battery pack !

e-beach said:

A question on your experience with recycled 18650 batteries...

I recently bought 49 dell laptop battery packs in 2 lots. One lot was 27 packs and the other 22 packs. The total price was just under $70 USA for all the packs.

I have yet to fully count the total number of batteries but it's about 325 used 18650's. Mostly Samsung 2.600 amp hour cells but also Sony's LG's and others I have yet to identify.

Right now, I have them separated into 3 groups.
A: Those that are above 3.25 volts,
B: Those that are 3.24 to 2 volts
C: and those that are 1.99 to zero volts.....I have about 113 of these.

At this point my question is what should I do with the 113or so that are 1.99 to 0 volts. They may be the dangerous ones and I am wondering if they are worth trying to revive with a slow charge to see if any are salvageable.

From your experience is it worth trying to salvage low voltage cells?

Thanks!

Click to expand...

Once again ... sorry!
Please tell us about your battery builds.

Click to expand...

e-beach said:

Yep, I asked that question....Last year :lol: ( Sun Nov 03, 2013 6:34 pm ) :lol:

Click to expand...

e-beach said:

- Nov 23, 2013
A question on your experience with recycled 18650 batteries...

I recently bought 49 dell laptop battery packs in 2 lots. One lot was 27 packs and the other 22 packs. The total price was just under $70 USA for all the packs.

Click to expand...

e-beach said:

- March 22, 2014
Ok, after reading this thread I made an impulse buy of what turned out to be about 325 18650 cells. They cost me $70.00 USA and were mostly from used dell laptop battery packs. I have separated all the cells and I am just finishing the study of potentially usable Samsung cells (just over 100 pink, 2600 mAh cells of various dates and chemistry)

Click to expand...

e-beach said:

- March 23, 2014
As for posting my methodology, thank you for the recommendation, I will when I am finished with the 325 cells I have for this test. I still have 80 or so cells to charge and discharge. As a side bar, I fully intended to post my findings in the first place. I just got caught up in responding to a question on your thread.

Click to expand...

Sounds like you are still testing your first batch of cells .. and have never built a pack!

If you are going to recommend drastic changes to the accepted, "Tried and true", cell testing methodology, I think it to be reasonable that you post up your qualifications.

Especially since multiple, much more experienced, members disagree with you ...

You seem to be belligerently oblivious* to the fact that capacity and self-discharge are two separate and largely unrelated factors.
Recommending that others not test for self-discharge is sabotaging pack construction and endangering others.
It is one matter to test your own theories ... but to actively promote that others use your, untried, warned against, methods is just nasty ... might even border on the criminal ... ?

I think it would be only responsible, that you be truthful ... and proclaim that you have no battery pack building experience and that you base your opinion on your haphazard and incomplete cell testing ... that has dragged on for more than 4 months.

* More accurate might be stubbornly ignorant

e-beach, self discharge is actually a factor and I personally took 3 packs apart because of this issue. Capacity loss and degradation is nothing in comparison and I am not taking possible internal short in account. Imagine a pack with 1 (or more)S constantly self discharging, there is no way to use it without constant balancing and LV monitoring. Sure, if BMS is used, you could take this gamble, but again, internal short seems a logical end of self discharging cell.

parabellum said:

e-beach, self discharge is actually a factor and I personally took 3 packs apart because of this issue. Capacity loss and degradation is nothing in comparison and I am not taking possible internal short in account. Imagine a pack with 1 (or more)S constantly self discharging, there is no way to use it without constant balancing and LV monitoring. Sure, if BMS is used, you could take this gamble, but again, internal short seems a logical end of self discharging cell.

Click to expand...

I never stated that "dud" cells were not a problem in a battery packs, I stated that through my thorough research I have found that the "bleed down" test was wholly inaccurate because "dud" cells don't always bleed down: a repeatable finding that seems to put the OP into shrill hysterical fits of mistruths, yelling and insults....rather entertaining actually. :lol:

As for my method, I have posted a portion of my findings with more to come .

@parabellum: But in all fairness to science, I am curious as to how you selected your cells for you pack build. What process did you use to determine which cells to use and which not to use?

e-beach said:

I never stated that "dud" cells were not a problem in a battery packs, I stated that through my thorough research I have found that the "bleed down" test was wholly inaccurate because "dud" cells don't always bleed down:

Click to expand...

Well. what is your problem then?
No one has ever stated that a bleed down test was to test for all types of "dud" cells!

Bleed down test is to test for self-discharging cells ... totally different from low capacity, and largely unrelated.

After re-reading your posts, it seems that either:

1. You have no understanding of what self-discharge is
Tho, I and others have repeatedly warned about and tried to explain this to you.

or

2. You mistakenly believe that "Because simply charging and then discharging a cell, or cells through a watt meter exposes all cell characteristics necessary in determining which cells to use in a safe pack."
Wrong ... wrong ... very wrong!
You might, figuring the odds, eliminate some, but capacity is not directly relateable to stable self-discharge!

I admit, when I built my 1st recycled Li-ion pack, back in 2008, I did not take self-discharge into consideration.
Of course, I was on my own, with no available instruction or tech support, I was the 1st, that I knew of, laptop cell recycler.
I had to develop all my own testing and building procedures, resulting in some needlessly sketchy early builds.
Self discharge, (bleed down), testing was implemented immediately after my 2008 builds demonstrated self unbalancing.
My more recent packs, with capacity matched banks and self discharge elimination have cycled hundreds of charge-discharge cycles and remained accurately balanced, at full and empty, within a couple 100ths V. Without any BMS! ... But I do monitor all bank voltages ... especially during preliminary cycles!

So I began posting up my experiences for others to benefit from.

But if you need to, can only, learn from your own mistakes ... you're welcome to.

Just stop posting for other newbies that procedures, developed, optimized and tuned over the past 7 years and dozens of builds are unnecessary and worthless.

DrkAngel said:

So I began posting up my experiences for others to benefit from.

Click to expand...

Wich we really appreciate!!

e-beach is also welcome to share his battery builds...

e-beach said:

parabellum said:

e-beach, self discharge is actually a factor and I personally took 3 packs apart because of this issue. Capacity loss and degradation is nothing in comparison and I am not taking possible internal short in account. Imagine a pack with 1 (or more)S constantly self discharging, there is no way to use it without constant balancing and LV monitoring. Sure, if BMS is used, you could take this gamble, but again, internal short seems a logical end of self discharging cell.

Click to expand...

I never stated that "dud" cells were not a problem in a battery packs, I stated that through my thorough research I have found that the "bleed down" test was wholly inaccurate because "dud" cells don't always bleed down: a repeatable finding that seems to put the OP into shrill hysterical fits of mistruths, yelling and insults....rather entertaining actually. :lol:

As for my method, I have posted a portion of my findings with more to come .

@parabellum: But in all fairness to science, I am curious as to how you selected your cells for you pack build. What process did you use to determine which cells to use and which not to use?

Click to expand...

1 pack was made of konions makita packs, and I did not used any method to determine self discharge, I just did capacity test of cell pairs. 1 pack was Turnigy 5Ah 4s hard case (pack made of multiple cases), I just kind of ignored slight V difference on arrival. 1 pack was 2 years in use (made of 5Ah 4s hard cases) then 1 P started to self discharge and my bulk charge/discharge procedure was to risky. (Last pack was not compressed i right way and puffed)

diggler said:

hi im just getting into this and I my opinion It makes no difference which test you do first. because a bad battery (meaning no capacity) apparently can sit for a month or so showing no sings of bleed down or internal short ill call it. But a battery that has a hi capacity % can still have an internal short....So I agree with both of you but more so drkangle cause either way the bleed down test will have to be done sooner or later. you can do the capacity test first it doesn't matter but then your just going to have to turn around and do a bleed test after that or vicea virca tell me if im wrong like I said im a week into this stuff and just reading about it but I do need to know cause I got 336 cell coming and only a imax b6 with parallel board and balance leads. so charging is going to be a pain! I like your way of rapid/balance charging with a bulk charger in parallel with a balance charger that has a diode. Genius! I never would have thought of that! So you put the diode on the bulk charger right? then set the volts a little bit lower than the final topping charge right? again please tell me if im wrong I have 0 experience with this but I did take automotive tech. so I have a basic understanding of it. I plan on doing the whole light bulb thing to capacity test but I think im going to get a watt meter also cause I will need it to measure how much capacity I have left on a bike ride anyway. I have many more questions and ideas but ill wait a bit.

Click to expand...

Yes, place the blocking or directional diode on the lower voltage bulk charger ... you do not want the balance charger bleeding back into the bulk charger.

I can not find-remember where I posted up this suggestion.
So let me more fully explain:

Using a large capacity bulk charger in conjunction with a balance charger

Many use a larger - cheaper bulk charger to quickly charge their batteries.
Others, for convenience or safety reasons, use a balance charger.

I say "why not use both, at the same time?"
If you set the bulk charger to a slightly lower voltage and add a blocking-directional diode, it can be run paralleled with a balance charger!

eg. Initial charging will use the full 7A of the bulk charger and the full 3A of the balance charger.
When voltage drops to the bulk chargers set limit it will cease contribution and the charging, and balancing, will be solely handled by the balance charger. (Diode is to isolate the bulk charger, preventing voltage feedback.)

This, effectively will give you a 10A balance charger for the cost of a 3A balance charger, a 7A bulk charger and a < 50¢ diode.

What would happen if i add one or two new cells per series group with higher C rate to a sanyo recycled battery build?

virtuhal said:

What would happen if i add one or two new cells per series group with higher C rate to a sanyo recycled battery build?

Click to expand...

Adding better surge rate cells, of compatible voltage, equally to each series should add capacity and surge capability.
I don't foresee any problem or downside.

In fact, I plan on building 9s8p laptop LiPo ... then adding 9s2p RC LiPo to improve capacity noticeably and surge capability greatly. Laptop and RC are both Lithium Cobalt formulations with similar voltage capabilities.

virtuhal said:

What would happen if i add one or two new cells per series group with higher C rate to a sanyo recycled battery build?

Click to expand...

You will significantly reduce the sag and make your capacity useful down to LVC, that means much more useful power. I actually paired LiMn with LiCo packs, no issue so far.

What if I use one cell that won`t charge above 4.17V? Will I still be able to charge the whole batterypack to 4.20?

kje said:

What if I use one cell that won`t charge above 4.17V? Will I still be able to charge the whole batterypack to 4.20?

Click to expand...

If charged to 4.20V, 4.17V cell will unbalance its bank, creating a pack requiring constant re-balancing.

Already recommended charging to only 4.10V.
3.6V cells seem better designed for 4.10V charge, via capacity map - for less rapid deterioration.
Charging to 4.20V will give you, possibly 10% more capacity but cut lifespan in ½.
¡Already recommended charging to only 4.10V!

Ebeach, the part you keep missing isn't that anyone is saying all bad cells have high self discharge, but rather that all cells with high self discharge are bad. It's a perfectly valid and simple battery test....charge them all up and let them sit. Those which don't hold their voltage well go directly into the recycle bin and require no further time consuming testing.

kje,
If a 3.6v-3.7v nominal cell can't be charged to 4.2v then consider it a dud. You're going thru a lot of work, so you only want the best and brightest in your primary pack, because all it takes is one bad one to kill them all, so thoroughly weed out those little terrorists. It should be easy to find lots more discarded 18650's, so instead of spending effort on questionable cells, use the time to round up lots more for testing. Then you can make a nice big pack of cream-of-the-crop cells.

but this is just not true. this guy angel is a bully for using the shouting words when ebeach gave him some good advice.

you are using drop in resting cell voltage as an indication of self discharge rates and you did that with absolutely no proof there is any connection between the two.

ebeach gave a good rationale for segregating cells based on the true storage capacity and this guy angel insulted him again about how he doesn't know what he is about because he doesn't build battery packs like angel does. but nobody builds batteries like this guy angel does.!!!!!!!!!!!!!!!!!!!!!

i never go onto this thread because the techniques are so strange and irrelevant to how a battery needs to be tested but for him to deliberately insult ebeach for trying to put a glimmer of truth into these mindless ambiguous statements is pretty annoying. who really believes this stuff anyway?

Everything on this forum needs to be qualified. The more ad hominum, the more qualification.

Whatever the method of testing and screening, don't forget that excess heat while soldering into the final pack can ruin cells, too. To monitor final pack configurations, I'd think thermal cameras would prove indispensible for identifying the spoilers. It would be interesting to see if thermal camera information correlated with any methods discussed in this thread.