Charging Lithium Ion in parallel

[FredM]

I am thinking of making a pack of 4 Li-Ions in parallel, if I charge them like this they will balance themselves even if I setup my charger to think it is charging a single Li-Ion correct.

For example say I make a pack of 4 1.0Ah cells all in parallel then tell my charger it is charging a Li-Ion of 4.0Ah at 3.7V.

They will even themselves out correct?

 

[VidPro]

yes, almost all the rechargable batts work fine in parellel, increasing the total amperage but not changing the voltage, giving you more "capacity" at the same voltage (just soes were clear on parellel).

with a normal li-ion charger there is a peak maximum voltage ,so this to works out pretty good, with a ni-mhy "smart" charger working off of voltage drop detection, it doesnt work as well. so li-ion works pretty darn good in parellel, because of the max voltage charging alogrythm.

how well connected the cells are would be the cause of any variations, so that is one concideration, using long small wire to connect the parellel set, the resistance of the wire during discharge or charge could become a factor. even with poor connection, in eventuality they balance out anyways, but if you want it to work as a big fat cell, then have big fat low resistance connections.

other possible conciderations are if one ages , or becomes dead, self discharges or heats on charging, that will effect the whole parellel grouping. this would only reveal itself in time or if one of the cells are problematic.
EX: 4 cheap junkey cells of which 1 out of 4 becomes bad , that cell will take on the charge in the form of heating that cell, and that cell will cause the "pack" to self discharge.

so when doing that, and using cheap or unreliable cells, the RATE of charge would be important, say charging 4x slow and taking 4times as long, or trying to charge the 4X 4Times as fast as ONE. If one cell goes bad, the 4times as fast of charging will increase the chances of thermal overload of the bad cell.

in ideal conditions the cells act as a team, acting like one Big fat cell, only if one decided to be junk on you (which is more often over TIME) then that cell could cause the whole pack issues beyond that cell malfunctioing, because that cell will "take the charge" when its bad.

and what i mean by take on the charge, i mean the voltage of that BAD cell wont rise, so when you hit the pack with 4c (say) all 4c will go to the one lower voltage bad cell that will not charge, the other cells will just be standing there watching the bad cell get all red faced and angry hot

note: i added in all the caveats, cause they CAN exist, not because they always will, its just good to know.

so a simple thermal fuse or thermal breaker is a Great thing to add to a parellel setup, and the thermal item should be capable of cutting out for ANY battery in the pack that would go bad, meaning should be contacting every cell or doubled up so every cell can have one.
You could also limit the max charge current to that of a single cell, in the pack, so if one cell should ever decide to go bad on you, the heating wont cause a thermal overload.

 

[LuxLuthor]

Yes this is perfectly fine to charge and discharge Li-Ion's in parallel...and the set in parallel will all be balanced. If you have a second set of parallel like 2s3p then they will also be balance charged, but they won't be balanced with the first set of parallel.

You cannot charge NiMH in parallel, but they can be discharged in parallel

 

[rizky_p]

Yes this is perfectly fine to charge and discharge Li-Ion's in parallel...and the set in parallel will all be balanced. If you have a second set of parallel like 2s3p then they will also be balance charged, but they won't be balanced with the first set of parallel.

You cannot charge NiMH in parallel, but they can be discharged in parallel

didnt know ni-mh can't be charged in paralel, good to know. No wonder my paralel 900mah AAA nimh doesnt reach 1800mah like it should when charged.

 

[LuxLuthor]

Yeah it's not as well known. I have a post about it with links in my battery making sales thread.

 

[VidPro]

didnt know ni-mh can't be charged in paralel,

potentially it can be done, just not done using the standard charging methods they use FOR the ni-mhy. one cell will v-drop , then it takes all the charge, while the yet uncharged one, doesnt get fully charged.

there are 2 methods of charge you can use to charge parellel ni-mhy, one is to be below the overcharge rate for a SINGLE cell, so even if you have 3 cells in parellel, you keep the rate at 1/10C for one , or 1/30c for a pack of 3. that way no cell item in the pack will ever v-drop. and you stay within "spec" for an overcharge rate , by staying within spec for the SINGLE cell. so its a slow charge, even slower still. this will work ok for something that Lives on the charger, or charge cradle or something.

the other method, is to use the same method that li-ion uses, use a voltage max, and slow charge neer the end. sorta a float type of charge like you would use for a lead acid, a locked out 1.4V per series cell CAN work for most stuff, but it wont always. because it wont ever reach an overcharge state, there will be no v-drop, so no cell will go down in voltage and start taking all the charge.
with a voltage max, you can fast charge at first, cause at first it will never reach the voltage max, and then as the voltage differential drops it goes slower and slower, so you can still get to an 80% charge pretty fast, just like you can with li-ion, then it slowly floats up to where the voltage difference between the charge and the battery have minimal current flow (topping it off).

|--1.35--|
|--1.50--|
in exageration, this parellel set one cell has v-droped, and as you can see it will be the one taking the charge your hitting the pair of them with.

both methods are not intuitive or standard (aka not understood well), but they work just fine, voltage max for ni-mhy is more of a feeling than a specific voltage, but with the right voltage for the battery , you can "float" the charge , with the wrong one, you get v-drop and it all goes to heck like before.
entirely possible, just could be concidered more of a "manuel" method, because its not going to be as easy to get a stupid machine to do it right for every possible type of cell item, or the way it reacts..

they DO use "smart" charge v-drop termination on some parellel ni-mhy, its done by a torturous fast speed, that is sure to incur a v-drop in all cells, even if one cell drops before the other, then the "smart" computer ends up detecting the first cells v-drop or misses it, and doesnt see a v-drop till BOTH cells vdrop, which has already now overcharged the one that v-dropped first. so when they use that method it "works" but it almost guarentees a hard overrate overcharge.

oops edited that, i meant to say the slow charge rate

 

[rizky_p]

why in a serial configuration a v-drop doesnt couse false charge termination, since (like in paralel) there are more than one batteries in the setup?

thanks

 

[marschw]

If you can't charge nimh in parallel, why do many of the cheap 4AA chargers only work with 2 or 4 cells? My assumption was that they were really only 2 separate channels, despite the 4 slots for AA cells in the charger; thus 2AA in parallel.

 

[VidPro]

why in a serial configuration a v-drop doesnt couse false charge termination, since (like in paralel) there are more than one batteries in the setup?

thanks

because in serial at least when one drops the whole string of thems voltage drops, giving minor notice that ONE has dropped, which doesnt mean the others are fully charged yet.
even in serial sometimes it will take 2 items to do a harder drop for a computer to begin spotting the drop, a good computer for serial charging of ni-mhy is more sensative to drop like noticing that the voltage is just not going UP anymore, then just starts topping off to finish at a low rate.

so it can cause false termination and is still hard for a machine to interpret the end points. any slow topping can still balance out the series set, even if it was fast charged initially.

in exageration again
-1.48-1.50-1.51- =4.49
reaches full charge and vdrop point for one cell
-1.49-1.51-1.45- =4.45 <- visable drop

so the whole packs voltage still has some visable drop when one starts dropping.

the problem still comes when the voltage of the others is still rising rapidly, when one in the series drops. usually occuring when there are MANY more in series.
-1.48-1.47-1.46-1.48-1.44-1.48-1.45-1.47- = 11.73
-1.50-1.48-1.48-1.47-1.46-1.50-1.47-1.48- = 11.84
as you can see in this exagerated example the red one peaked and was vdroped and dropping hard even, yet the voltage on all the others was still doing what it does, going up, making the single v-drop in there totally unnoticable.
and your usual series "smart" charger will notice what is going on more around here
-1.50-1.50-1.50-1.47-1.48-1.47-1.48-1.47- = 11.87
which is right after it has slightly hurt a few of the cells, but still close enough for the pack to work ok for many cycles, just not Ideal.

in both series and parellel arangements , stuff can still happen that makes "smart" ni-mhy charging hard to be a perfect thing, the "newer" li-ion technology by using its charge "Method" which does not overcharge to detect a end to charge , is a better method.
i cant find any reason why the same methods (for li-ion) cannot be applied to ni-mhy, even if they relentlessly insist on reaching the v-drop point for charging ni-mhy, its not nessisary to use that method to charge ni-mhy, or to balance them out either.
we have sucessfully used voltage max, and slow current control to charge series & parellel and BOTH series parellel ni-mhy configurations for years, ok with a few problems with knowing how to set the voltage or current but once i get that figured all perfectally it works repeatedly.

 

[VidPro]

If you can't charge nimh in parallel, why do many of the cheap 4AA chargers only work with 2 or 4 cells? My assumption was that they were really only 2 separate channels, despite the 4 slots for AA cells in the charger; thus 2AA in parallel.

usually those chargers that require 2 are (usually) doing so in series, often there is a MAX voltage of the series channel (~3.0), and a slower rate. they do ok as long as the batts were used as a team, it gets pretty bad when the batts are totally wacked different discharge. the OLDer ones were such a slow rate it wasnt an issue at all, some of the newer ones the rate is a bit higher, but still sorta a lower rate.

 

[rizky_p]

Thanks.

because in serial at least when one drops the whole string of thems voltage drops, giving minor notice that ONE has dropped, which doesnt mean the others are fully charged yet.
even in serial sometimes it will take 2 items to do a harder drop for a computer to begin spotting the drop, a good computer for serial charging of ni-mhy is more sensative to drop like noticing that the voltage is just not going UP anymore, then just starts topping off to finish at a low rate.

so it can cause false termination and is still hard for a machine to interpret the end points. any slow topping can still balance out the series set, even if it was fast charged initially.

in exageration again
-1.48-1.50-1.51- =4.49
reaches full charge and vdrop point for one cell
-1.49-1.51-1.45- =4.45 <- visable drop

so the whole packs voltage still has some visable drop when one starts dropping.

the problem still comes when the voltage of the others is still rising rapidly, when one in the series drops. usually occuring when there are MANY more in series.
-1.48-1.47-1.46-1.48-1.44-1.48-1.45-1.47- = 11.73
-1.50-1.48-1.48-1.47-1.46-1.50-1.47-1.48- = 11.84
as you can see in this exagerated example the red one peaked and was vdroped and dropping hard even, yet the voltage on all the others was still doing what it does, going up, making the single v-drop in there totally unnoticable.
and your usual series "smart" charger will notice what is going on more around here
-1.50-1.50-1.50-1.47-1.48-1.47-1.48-1.47- = 11.87
which is right after it has slightly hurt a few of the cells, but still close enough for the pack to work ok for many cycles, just not Ideal.

in both series and parellel arangements , stuff can still happen that makes "smart" ni-mhy charging hard to be a perfect thing, the "newer" li-ion technology by using its charge "Method" which does not overcharge to detect a end to charge , is a better method.
i cant find any reason why the same methods (for li-ion) cannot be applied to ni-mhy, even if they relentlessly insist on reaching the v-drop point for charging ni-mhy, its not nessisary to use that method to charge ni-mhy, or to balance them out either.
we have sucessfully used voltage max, and slow current control to charge series & parellel and BOTH series parellel ni-mhy configurations for years, ok with a few problems with knowing how to set the voltage or current but once i get that figured all perfectally it works repeatedly.