Mixing 18650 capacities

Bobbych

Member
Joined
Feb 18, 2020
Messages
30
Hi All

So I have over 2000 used 18650 cells. I have capacity tested them all and now have a range of between 1200-2000mAh.

I'm looking to create the biggest 24v capacity setup I can with these cells. 7S???P.

Im still leaning but Im guessing mixing all those capacities (Using rePackr) would not be a good idea, or is it?

Assuming it is not a good idea what is the maximum capacity range I should be using? 2000mAh down to ????

Thanks in advance
 
If it wasn't a good idea, why would we have a link to use it????

You can mix any combination of cell capacities with any range.
The thing you need to take into consideration is what was the cells capacity at the start of its life, and what is it now. If it's more than about 25% drop, then you probably don't want to use it.

Original 3000mAh
Now 1700mAh
Difference 56%

Original 2000mAh
Now 1700mAh
Difference 85%

As you can see, the 2nd cell is much better condition than the first, even though they have same Now capacity. So it is highly recommended to compare "all" cells against their original specs
 
Korishan said:
If it wasn't a good idea, why would we have a link to use it????

You can mix any combination of cell capacities with any range.
The thing you need to take into consideration is what was the cells capacity at the start of its life, and what is it now. If it's more than about 25% drop, then you probably don't want to use it.

Original 3000mAh
Now 1700mAh
Difference 56%

Original 2000mAh
Now 1700mAh
Difference 85%

As you can see, the 2nd cell is much better condition than the first, even though they have same Now capacity. So it is highly recommended to compare "all" cells against their original specs
ThanksKorishan. My line saying its not a good ideawas referring to mixing the cells not rePackr.
Anyway. my cells were originally2100 and 2000 so i will exclude any below 1600mAh. (20% Loss)

Thanks Again
 
Mixing cells depends. If you make sure the weakest cell is the One determening max useage its fine.

Make sure to mix so you get even Number for each type in each pack. (Repackr wont do that....)

Them you get the best chance for it to last
 
Here is what You can do. If You have 2100 cells You can make 3 x 7 banks each 100 cells.

Prepare 21 cell holders 100P and place them on a table. (Quite a large table ;) ) Name the Banks 1 - 21.

Sort all Your cells in decending order. Start placing the highest 21 cells. Highest cell goes in Bank 1, Next highest in Bank 2, and so on until You have placed cell number 21 in bank 21. Cells 22 to 42 is placed in Bank 21 - 1, so the lowest is placed in Bank 1.
Next batch of 21 is placed 1 - 21. Next batch in 21 to 1. Continue this sequense until You run out of cells or all banks are full.

By doing it this way, each bank will be equal to the next. By building 21 banks in one go, You will end up with 21 banks equal in capacity and with the maximum capacity possible with Your batch of cells.

Best of luck :)

ChrisD
 
Interesting. Would I connect sets of 3 in parallel the the backs of 7 in series?
 
You can connect banks 3 in parallel and then 7 in series, this gives You 7S300P, or You can connect 7 in series and then parallel the 3 resulting 7S100P.

7S100P x 3 will need 3 BMS boards. 7S300P will need only one, but significantly larger BMS.

Say You want to draw 75 Amps (3 KW at 25 Volts) each 7S100P will work with a 30A+ BMS, and they can be found at a reasonable price. 7S300P will require at least a 80A+ BMS, and You may find that this is going to cost You more.

I would go for 100P, but the choice is Yours.:)

ChrisD
 
ChrisD5710 said:
You can connect banks 3 in parallel and then 7 in series, this gives You 7S300P, or You can connect 7 in series and then parallel the 3 resulting 7S100P.

7S100P x 3 will need 3 BMS boards. 7S300P will need only one, but significantly larger BMS.

Say You want to draw 75 Amps (3 KW at 25 Volts) each 7S100P will work with a 30A+ BMS, and they can be found at a reasonable price. 7S300P will require at least a 80A+ BMS, and You may find that this is going to cost You more.

I would go for 100P, but the choice is Yours.:)

ChrisD

This is a great question and answer and I think would be a useful explanation, particularly in some form in the FAQ, when it comes to pack design, system setup, etc. The systems above 'sound' the same but even in this simple explanation it's clear there are considerations that, I believe, many are not inherently aware of when thinking about the design of their systems. Especially for people building 'smaller' systems, systems that are not always going to continue to grow (I think systems of that nature tend to be 'easier' to plan because you're always adding stuff).

To highlight some of the more important elements, from the aspect of the size of the system (3kw system in the above example from a 24V battery) and understanding the ramifications and how they, directly, impact system setup. Things like BMS requirements as well as wiring/busbars (see below).

Also, correct me if I am wrong, but going the 3 x 7s100p system would also allow smaller gauge wire/busbars for the 100p's versus going the 7s300p route? I am ONLY discussing the wire/system for the busbars for the 100p or 300p packs ... I understand that when doing the 3 x 7s100p system, you will still ultimately end up with the same requirements for wire gauge to draw the 75amps, for example going to a 3000w inverter.
 
daromer said:
Mixing cells depends. If you make sure the weakest cell is the One determening max useage its fine.

Make sure to mix so you get even Number for each type in each pack. (Repackr wont do that....)

Them you get the best chance for it to last

Yes the SLSrePackr won't do that but the NemosRepackr does.
You do have to have M$ Excel but if you have your cells numbered it will create the packs for you with the best ballance and the most Ah.

After the Repacker is done it will assign a pack number to each cell then you can sort by cell identifier (number) and you can easily find the cell and place it into the corresponding pack.



image_nmbuif.jpg

Wolf
 
I am a bit confused about this thread. Seems like you are all saying that mixing different capacity batteries in parallel is ok??
If I have 3000mAh 2500mAh 2000mAh acting as one "cell", thus in parallel, shouldn't the 2000mAh be charged full before the other 2? Thus, it will either be overcharged or it will provide current to the other 2 batteries and can therefor create huge current flow to them? Also, is really mixing batteries with different resistance a good ide? That would mean the batteries will all charge in different speeds. I am building a pack myself and is also looking for the answer to "is it ok to mix different capacities/resistances in parallel".
 
I am a bit confused about this thread. Seems like you are all saying that mixing different capacity batteries in parallel is ok??
If I have 3000mAh 2500mAh 2000mAh acting as one "cell", thus in parallel, shouldn't the 2000mAh be charged full before the other 2? Thus, it will either be overcharged or it will provide current to the other 2 batteries and can therefor create huge current flow to them?
It's important to understand that when you parallel cells - they ALL have exactly the same voltage all the time. You can't overcharge 1 cell or undercharge another cell when they are in parallel. 3000 to 2500 to 2000 is not a huge difference given adequate connections between the cells and the buss. If you have 3ah cells and 2ah cells a bit more current will flow in/out of the 3ah cells compared to the 2ah cells over the discharge cycle. I don't know the math - but from experience I can attest its not a issue.

In my case, my battery is designed to operate at <200ma/cell. I have batteries at 2000ma/cell and 2600ma/cell and 3000ma/cell (roughly equal distribution) and there's been no noticeable affect in balance or degradation. The 2000mah cells have just reached 1,160 cycles as of yesterday.


Also, is really mixing batteries with different resistance a good ide? That would mean the batteries will all charge in different speeds. I am building a pack myself and is also looking for the answer to "is it ok to mix different capacities/resistances in parallel".
Again, in parallel, the cells MUST be the same voltage all the time. So they don't charge at different speeds. However, the higher the resistance, the warmer a cell can get depending on the level of current and it can shorten their lives if its out of spec.

In my batteries, my IR varies (roughly) between 20mOhm and 80mOhm and I charge/discharge <200ma/cell (low stress) and its not a detectable issue. The cells do not show any warming at all that I can tell - regardless of 2000mah or 30000mah.

I did have a set of cells with 250mOhm IR and these were so out of whack with the others I had to pull the packs

@Wolf has done extensive research & documentation on acceptable ranges of IR per cell types - and found that part of testing cells should include an IR check so you avoid resistance issues completely by using healthy cells :)
 
Last edited:
It's important to understand that when you parallel cells - they ALL have exactly the same voltage all the time. You can't overcharge 1 cell or undercharge another cell when they are in parallel. 3000 to 2500 to 2000 is not a huge difference given adequate connections between the cells and the buss. If you have 3a cells and 2a cells a bit more current will flow in/out of the 3a cells compared to the 2a cells over the discharge cycle. I don't know the math - but from experience I can attest its not a issue.

In my case, my battery is designed to operate at <200ma/cell. I have batteries at 2000ma/cell and 2600ma/cell and 3000ma/cell (roughly equal distribution) and there's been no noticeable affect in balance or degradation. The 2000mah cells have just reached 1,160 cycles as of yesterday.



Again, in parallel, the cells MUST be the same voltage all the time. So they don't charge at different speeds. However, the higher the resistance, the warmer a cell can get depending on the level of current and it can shorten their lives if its out of spec.

In my batteries, my IR varies (roughly) between 20mOhm and 80mOhm and I charge/discharge <200ma/cell (low stress) and its not a detectable issue. The cells do not show any warming at all that I can tell - regardless of 2000mah or 30000mah.

I did have a set of cells with 250mOhm IR and these were so out of whack with the others I had to pull the packs

@Wolf has done extensive research & documentation on acceptable ranges of IR per cell types - and found that part of testing cells should include an IR check so you avoid resistance issues completely by using healthy cells :)
Ok, I guess it makes sense. However, when charging, the charger will give something like 4.35v to the 3 batteries in parallel. Thus, current flows from the charger to the 3 cells. If one cell has lower capacity than the others, logic says that it should reach 4.2v faster than the others. If it is like you say, that all the batteries stays at the same voltage, that would mean that the low capacity cells charges SLOWER that the others, thus taking less current. That would mean that the higher capacity cell would get more current from the charger, which could damage it. Because all 3 cells just cannot take the same amount of current and still reach full capacity at the same time. Something must give.
 
I am building a pack myself and is also looking for the answer to "is it ok to mix different capacities/resistances in parallel".
To a certain degree.
Short answer stick to a max mAh difference of 400 between cells and try to keep the IR difference to <15mΩ.
If I have 3000mAh 2500mAh 2000mAh acting as one "cell"
That is correct they will be acting as ONE cell. The voltage will be the same on all cells so charging will occur according to voltage not capacity of the cell. You cannot overcharge any cell as long as the pack voltage does not go over 4.2V.
Water in a vessel is an excellent representation of voltage and capacity. It is a crude representation nevertheless you can see in the illustration below we have 4 separate vessels of different capacities. They are being filled (charged) by the Tank. The water (voltage) will seek its own level in all containers. This is how cells will react in parallel with different capacities.
1635861014458.png
If all you have is a conglomerate of different manufacturers / part numbers of cells then in my experience your pack voltages below 3.6V will begin to deviate a bit between packs. In my case the max deviation is about 0.09V This is not a huge concern as long as you monitor it and keep track of any irregularities.
I built my first 14s80p from basically 4 manufactures. Sanyo, Panasonic, LG, and Samsung with 28 different part # as shown here with min, max, and average values.
1635864187161.png
The result was a reasonable Frankenstein pack that performs quite well. Here is a trace of the pack voltage and deviation voltage per pack through the charge and discharge cycles.
As you can see once voltage drops to ≈ 3.6V the pack voltages start to seperate a bit. The purple tracer shows that.
1635863852938.png
My 2nd battery packs are all of the same cell and manufacturer with a mAh difference of ≈300 and an IR difference of <5mΩ
1635864928764.png

The results of closely matched cells is very apparent by the voltage difference between packs during charging and discharging.
In this case < 0.015V through the charge and discharge cycle.
1635863808731.png
Additional information can be found in my thread of how batteries act in parallel.
Wolf
 
To a certain degree.
Short answer stick to a max mAh difference of 400 between cells and try to keep the IR difference to <15mΩ.

That is correct they will be acting as ONE cell. The voltage will be the same on all cells so charging will occur according to voltage not capacity of the cell. You cannot overcharge any cell as long as the pack voltage does not go over 4.2V.
Water in a vessel is an excellent representation of voltage and capacity. It is a crude representation nevertheless you can see in the illustration below we have 4 separate vessels of different capacities. They are being filled (charged) by the Tank. The water (voltage) will seek its own level in all containers. This is how cells will react in parallel with different capacities.
View attachment 26399
If all you have is a conglomerate of different manufacturers / part numbers of cells then in my experience your pack voltages below 3.6V will begin to deviate a bit between packs. In my case the max deviation is about 0.09V This is not a huge concern as long as you monitor it and keep track of any irregularities.
I built my first 14s80p from basically 4 manufactures. Sanyo, Panasonic, LG, and Samsung with 28 different part # as shown here with min, max, and average values.
View attachment 26404
The result was a reasonable Frankenstein pack that performs quite well. Here is a trace of the pack voltage and deviation voltage per pack through the charge and discharge cycles.
As you can see once voltage drops to ≈ 3.6V the pack voltages start to seperate a bit. The purple tracer shows that.
View attachment 26403
My 2nd battery packs are all of the same cell and manufacturer with a mAh difference of ≈300 and an IR difference of <5mΩ
View attachment 26405

The results of closely matched cells is very apparent by the voltage difference between packs during charging and discharging.
In this case < 0.015V through the charge and discharge cycle.
View attachment 26402
Additional information can be found in my thread of how batteries act in parallel.
Wolf
Sure, but if the cells can only have 400mAh difference and <15 ohm difference, it's pretty much the same cells! When dealing with salvaged laptop batteries, there is no way in hell you can keep those parameters :) I got about 300 batteries from laptops and planning to build a 250+ battery pack. Thus, I have to use what I have. We are talking anything from 1000mAh and up to 2500mAh with vastly different resistance.

To get what you describe, someone would need to get something like 1000 laptop cells to get 250 cells that works within your parameters. Thus, you have 750 batteries wasted. Nobody needs that many "flash light" batteries :) Then it would probably even be cheaper to just buy 250 new batteries in the first place.
 
Sure, but if the cells can only have 400mAh difference and <15 ohm difference, it's pretty much the same cells! When dealing with salvaged laptop batteries, there is no way in hell you can keep those parameters :) I got about 300 batteries from laptops and planning to build a 250+ battery pack. Thus, I have to use what I have. We are talking anything from 1000mAh and up to 2500mAh with vastly different resistance.

To get what you describe, someone would need to get something like 1000 laptop cells to get 250 cells that works within your parameters. Thus, you have 750 batteries wasted. Nobody needs that many "flash light" batteries :) Then it would probably even be cheaper to just buy 250 new batteries in the first place.
I salvaged Laptop batteries, aging between 16 and 1 years of age. I got around 600 in total, 50 dead, 450 within the >70% capacity left margin. 50 under it.
Wouldn't call that unworthy.
 
I salvaged Laptop batteries, aging between 16 and 1 years of age. I got around 600 in total, 50 dead, 450 within the >70% capacity left margin. 50 under it.
Wouldn't call that unworthy.
And those 70% only had 400mAh difference from the lowest capacity to the highest as well as resistance within 15ohms from each other?? I doubt it! That's the requirements he had. This is what the whole thread is about. If you can mix different capacities, like 1500mAh with 2500mAh. 400mAh difference is very narrow. 15ohm difference in resistance as well.
 
And those 70% only had 400mAh difference from the lowest capacity to the highest as well as resistance within 15ohms from each other?? I doubt it! That's the requirements he had. This is what the whole thread is about. If you can mix different capacities, like 1500mAh with 2500mAh. 400mAh difference is very narrow. 15ohm difference in resistance as well.
Original 3100mAh
Now 2200mAh
Left 71%

If your margin is 70% or higher, you can still use the cell. If you had one that even scored the whole 3100mAh and has 100% left, you can still use it.

Original 2000mAh
Now 1450mAh
Left 72%

Again, if margin 70% or more, you can still use the cell, even in parallel with a 100% 3100mAh one.

Maybe you should read @Wolf post again.

Quote: "Short answer stick to a max mAh difference of 400 between cells and try to keep the IR difference to <15mΩ"
Sorry i accedentally send a reply without having it written fully.
Also, he said "To a certain degree." Not "Stick to it 100% or else you'll die with a 100% chance"
 
To get what you describe, someone would need to get something like 1000 laptop cells to get 250 cells that works within your parameters.
Yep thats what needs to be done if you want a long lasting battery harvested from "laptop batteries".
In my case it took over 10,000 cells to come up with 1120 cells for my 14s80p Frankenstein Battery.
6000+ cells recorded in this spread sheet

After learning that IR is a perfect indicator of a cells SOH (State of Health) I was able to discard 4000 + (to the recycler) and giveaway a bunch of marginal cells as they would not (as you say) fit my criteria. I also saved a bunch of time by not testing these out of range /criteria cells.
I am not telling you how to build your battery as you can do whatever you like but you asked the question..............
I am building a pack myself and is also looking for the answer to "is it ok to mix different capacities/resistances in parallel".
So if you want to build a successful battery that requires little to no maintenance try to stay within the parameters. If this is not possible at least try to keep you cells SOH to > than 80% and avoid high IR >65mΩ cells. Which if you are pretty much using laptop cells( ICR chemistry) then that IR reading will be the very top margin for good cells as pretty much anything over 65mΩ will have a SOH below 80%. Except the occasional LG**S31865 which can have a reasonable result at 80mΩ The Sony US18650GR G* series cells also can have a very high IR but they do not play well with others so I would stay away from them.
These are my recommendations learned from experience, experiments and many other long evenings researching this.
My Frankenstein battery consists of these cells and their rated capacity as you can see between the highest and lowest there is an 800 mAh difference.
1635871229313.png
This is not tested capacity as by now you surely understand that most used cells will not give you 100% of rated capacity.
That being said if you take a 3000mAh cell at 80% SOH this = 2400mAh, likewise if you have a 2600mAh cell at 80% SOH it is 2080mAh you are within the 400mAh difference so it is not impossible. Does it take a little more work, some more cells, sure. Do you want to build a good battery or something that needs to be serviced on a regular basis. It matters none to me you are welcome to take the advice as you wish.

Best
Wolf
 
Back
Top