Connecting different-sized lithium batteries in parallel.

[xani]

Hi,

I have a bunch of batteries (from recycling used laptop batteries) that are fine (tested for excessive self discharge and such), same type but of varying capacity (say ~20-30% difference). I want to connect around 6 of them.

Due to nature of use (it's a backup power supply for rPI) I do not really care about max current or getting full capacity out of every cell.

My gut tells me they will just charge/discharge proportionally to their capactity and as long as I match voltages to avoid battery-to-battery flow during connecting it should be fine.

Am I missing something ?

[Wimberleytech]

Hi,

I have a bunch of batteries (from recycling used laptop batteries) that are fine (tested for excessive self discharge and such), same type but of varying capacity (say ~20-30% difference). I want to connect around 6 of them.

Due to nature of use (it's a backup power supply for rPI) I do not really care about max current or getting full capacity out of every cell.

My gut tells me they will just charge/discharge proportionally to their capactity and as long as I match voltages to avoid battery-to-battery flow during connecting it should be fine.

Am I missing something ?

I would not put them in parallel unless they were mounted in a fireproof box!

[xani]

Hi,

I have a bunch of batteries (from recycling used laptop batteries) that are fine (tested for excessive self discharge and such), same type but of varying capacity (say ~20-30% difference). I want to connect around 6 of them.

Due to nature of use (it's a backup power supply for rPI) I do not really care about max current or getting full capacity out of every cell.

My gut tells me they will just charge/discharge proportionally to their capactity and as long as I match voltages to avoid battery-to-battery flow during connecting it should be fine.

Am I missing something ?

I would not put them in parallel unless they were mounted in a fireproof box!

And why exactly? They were in parallel to begin with, just in pairs of 2

[mariush]

Can you sort them out by capacity and then use 3 of the lower capacity in parallel, and then in series with two of the higher capacity in parallel ?
Use a switching regulator to step-down to 5v.

[xani]

Can you sort them out by capacity and then use 3 of the lower capacity in parallel, and then in series with two of the higher capacity in parallel ?
Use a switching regulator to step-down to 5v.

Can't, the device I'm using it with (UPS Pico, a raspberry Pi ups board) is the one doing the charge/discharge and it supports only single cell work.

[digsys]

We've done hundreds of combinations of building left-over packs into parallel blocks, with never an issue (ok 1 or 2, but that's when we've pushed the envelope :-) )
I put fusible links from each cell to the "bus", usually I use a polyfuse, but it can be a single break at ~2x the max current you expect, and optionally one temp sensor somewhere where it can get an "idea" of any excess heat. That's not critical, and I often rely on VI profiles to let me know if something is drifting.
The MAIN thing I'd suggest is - keep max currents to under 1C of the pack, except for say 1-2 sec bursts max. If there is a large variation in capacities / condition, I'd reduce the max C limit to say 0.5 - 0.25C. As you suggest, the "better" cells will do most the work anyway.

[xani]

We've done hundreds of combinations of building left-over packs into parallel blocks, with never an issue (ok 1 or 2, but that's when we've pushed the envelope :-) )
I put fusible links from each cell to the "bus", usually I use a polyfuse, but it can be a single break at ~2x the max current you expect, and optionally one temp sensor somewhere where it can get an "idea" of any excess heat. That's not critical, and I often rely on VI profiles to let me know if something is drifting.
The MAIN thing I'd suggest is - keep max currents to under 1C of the pack, except for say 1-2 sec bursts max. If there is a large variation in capacities / condition, I'd reduce the max C limit to say 0.5 - 0.25C. As you suggest, the "better" cells will do most the work anyway.

That's the plan, even the peak current draw will be way below 0.25C (it is supposed to work several hours on a pack). Fusible link ain't a bad idea, original metal strip is way too thick anyway. Thanks.

[ConKbot]

If you're being reasonably gentle with the cells, paralleling is fine. If you're drawing heavy current, then it it can start being a problem. I.e. A 5 miliohm 2AH powertool cell in parallel with a 50 miliohm 3.5AH laptop cell in parallel getting hit with a 20A load won't discharge equally, with the power tool cell taking the brunt of the load at first, discharging, and the load gradually transferring over to the laptop cell. When you let the load off, then they will equalise, but if the balance is significantly off, you're now charging the hot power tool cell pretty rapidly as the terminal voltage on the laptop cell tries to recover. Not good.

Some fusing is still reccomended though, as one cell failing with the energy of the others being pumped into it, in addition to its own isn't great. In theory, as one is failing, the current interrupt device under positive terminal should open.  But ideally the fuse should blow before the CID, as the CID isn't going to trip till the cell had a good amount of pressure inside, and is near venting, or in the process of venting.

[JustMeHere]

It might be a good idea to put a well matched 1 ohm resistor on each battery lead.  But I really don't know what I'm talking about, so anyone else who knows better; please chime in. 

The idea is to eliminate variances in the ESR of the batteries.

[Rick Law]

It might be a good idea to put a well matched 1 ohm resistor on each battery lead.  But I really don't know what I'm talking about, so anyone else who knows better; please chime in. 

The idea is to eliminate variances in the ESR of the batteries.

Okay, I will chime in....  First, I must point out I am merely a hobbyist.  My solution has limitation and issues commensurate with my limited experience.

I have used recovered 18650 in parallel feeding a booster to feed 5V - typically use to recharge a phone via USB.  My phone draw just around 1.1A max.

I considered the 1ohm resister solution and rejected it.  At 1Amp, the resister will drop 1Volt.  That is too much loss for me.  Besides, it does not solve the issue of certain battery issues such as dead-short battery.  Given my limited experience, I do not have a big mental tool bag to draw from.  I decided to use SCHOTTKY diode.  I choose 1N5817 because I have them and I like them (comfortable using it and familiarity with it).

My setup has two slots for 18650, but I can use just a single 18650 or use two.  Each 18650 connects to 3x 1N5817 (3 diodes in parallel) to the "power bus" supplying Vin of the booster. (See attached photo, the switch controls if I am using 1x18650 or 2x18650 or off.)

Here is my thought process:

Since the diode has Vf (voltage forward is voltage-drop) that increase based on current, and lower Vf = less loss.  So, if I use multiple diode to split the current, I will be dropping down the Vf reducing loss.

For simplicity, I start my consideration with 2x 1N5817 in parallel.  Manufacturing variation means likely one will be better than the other.  More current will flow via the better one (lower Vf).  As more current flows, the Vf increases.  At some point, it matches the not-as-good 1N5817.  So, the 2x 1N5817 "self-adjust" letting more current go the lower Vf path, then the one with higher Vf kicks in to share the load.  After some experimentation, I found 3x 1N5817 was about the best balance keeping Vf at around 0.27V to 0.3V (average).  I do get more (smaller Vf) but not much more benefit going to 4x and beyond.

So, each of my 18650 flows to the power bus via 3x1N5817.  Similar logic applies to batteries when the batteries are out of balance.  The better 18650 gives out higher voltage and more current, so higher (3x diode average) Vf drop - until the voltage matches that of the lesser 18650 with its (3x diode average) Vf drop and it kicks in to help power the load.  So more current is drawn from the better battery but eventually both batteries contributes.  In most cases, the batteries became equal voltage in about 10 minutes.

With them connected via diode, current from the better one will not flow into the lesser one.  I am not sure how often a battery can died and become dead-short, but even if it does, I need not worry.

While I used only two (18650+diodes) in parallel, the same method can be done for more (18650+diodes) set in parallel to increase the total capacity.

That is my solution I have been using for a while.  Works well for me but I am not experience enough to determine if there is a deadly trap in there or not.  May be this "design" will benefit your thinking process in some way.

Edit: Reworded a few places to added the words "in parallel"...  The context is probably clear to most that the diodes are in parallel but may not be clear for everyone.

[ogden]

Resistors or diodes? - No need. Look at Tesla battery pack for example, there's only fuse for each cell in the bus. Before connecting make sure to charge all cells to exactly same voltage or balance them using resistors first, only when every cell has same voltage - replace resistor with fuse. Never ever directly connect discharged cell to charged one in parallel.

[Rick Law]

Resistors or diodes? - No need. Look at Tesla battery pack for example, there's only fuse for each cell in the bus. Before connecting make sure to charge all cells to exactly same voltage or balance them using resistors first, only when every cell has same voltage - replace resistor with fuse. Never ever directly connect discharged cell to charged one in parallel.

Yeah, when they are in at same voltage (and in good shape), resistors (or diodes) are unnecessary.  But the OP asks about connecting different-sized lithium batteries in parallel.  Diode/Resistors can be a way to allow them to be connected at random unbalanced state and provide some limited protection.

[ogden]

Diode/Resistors can be a way to allow them to be connected at random unbalanced state and provide some limited protection.

Note that battery pack cells for decades are connected in parallel using just safety elements, usually fuses. All you need to do - charge cells before pack assembly. Adding lossy resistors or diodes to battery pack just because you are lazy, ignorant or just ill-informed is strange solution (to say it politely).

[Rick Law]

Diode/Resistors can be a way to allow them to be connected at random unbalanced state and provide some limited protection.

Note that battery pack cells for decades are connected in parallel using just safety elements, usually fuses. All you need to do - charge cells before pack assembly. Adding lossy resistors or diodes to battery pack just because you are lazy, ignorant or just ill-informed is strange solution (to say it politely).

Mostly ignorance (which is a superset of ill-informed).  Since we are doing it as hobby, laziness would not be the issue.

For whatever reason, I am fearful a bad battery becoming a dead-short for other batteries to drive current through.  But I said that in my earlier reply already.

Both OP and I were talking about recovered laptop batteries.  In the process of recovery, it was visible how they were connected originally.  We are both finding ways to make the best use of those batteries after we cut it out of the pack.  I am sharing my limited experience in the hope that we can both learn more.  I surely would like to learn a better/safer way of "joining" randomly picked 18650's so I can use whichever fully charged 18650 cell(s) at hand.

Most of us have knowledge limitation.  We just have to work around it until we learn more.  The alternative is to sit on our hands and do nothing until we learn everything about a particular subject - which would be rather impossible.  I bet you by the time we have fully functional giga-bit quantum computer, most of quantum mechanics would still be mysteries even to the creators of the giga-bit quantum computer.

In my view, an ability to work around ones limitation is a good quality to have.

[ogden]

I am fearful a bad battery becoming a dead-short for other batteries to drive current through.

Oh yes indeed you shall be fearful. 18650 cell can supply enough energy to set everything around on fire. You need fuse for sure. Diode or resistor possibly can protect against fire accident, yet they both are less efficient as fuse - due to voltage drop. What's the point of using diode or resistor as fuse when you can use just fuse?

[Gallardo]

What I need to mention is that these batteries are used, the number of times of charging is different, the ESR is different, and the capacity is different. From the perspective of safety and self-loss, even if the diodes are added, they cannot be used in parallel because this is too dangerous.

My suggestion is that each battery is equipped with a booster with charging function and finally connected in parallel on the USB side. Each battery is independently charged and discharged, and there is reason to believe that it will be safer.

[digsys]

Theories are fine, but after many years working on EV battery packs, this is what actually happens -
The lowest ESR cell will supply a slightly higher % of the pack energy until it starts to match the pack "average" ESR, at which point they all become app equal -
Even with wildly different ESRs, the same process occurs, but now you need to factor in that the lower ESR cells will supply a larger share for most the capacity -
In this case, you would not let the capacity drop to what you would "normally" set. Also, derate the max C figures to suit worst case possibility for each cell. 10-50%?
IF you keep this in mind, then in nearly ALL cases that I have EVER seen, when a cell goes "bad", it goes high(er) impedance or O/C. They RARELY go short cct - usually only if you have over-driven them or subjected them to higher temps.
We will OFTEN condition / match up a pack by cycling it from ~ 3.5V <> ~ 1.5V (or less) over 2-3 cycles. Cells in parallel do a great job of conditioning each other. YMMV

[ogden]

From the perspective of safety and self-loss, even if the diodes are added, they cannot be used in parallel because this is too dangerous.

Not more dangerous than using fresh batteries/cells. Please do not spread misinformation.

[radiolistener]

But for safety purposes it's better to have good fire extinguisher during experiment 

[Rick Law]

I am fearful a bad battery becoming a dead-short for other batteries to drive current through.

Oh yes indeed you shall be fearful. 18650 cell can supply enough energy to set everything around on fire. You need fuse for sure. Diode or resistor possibly can protect against fire accident, yet they both are less efficient as fuse - due to voltage drop. What's the point of using diode or resistor as fuse when you can use just fuse?

Typically, rather than a fixed "soldered-together" pair of 18650, I "randomly select" out of the charged-battery box a pair of 18650 for use.  One may be better charge than another.  Worst yet, and this is the mistake I think I would make eventually, I might have a partially (or hardly) charged battery while the other is fully charge.

When they are parallel-connected via diodes, the diodes serves to prevent the lesser charged battery receiving an inrush of power/current from the better charged one.  It allows somewhat of an orderly sharing of the power.

[KL27x]

You have to derate the max discharge/charge rate in addition to capacity. As long as you are paralleling them to increase capacity rather than max charge/discharge rate, and your device has a relatively low draw compared to the cells' capabilities, then I wouldn't worry about it too much.

If you're stacking them together to increase discharge/charge rate, then your results won't be very reliable.

If a single cell doesn't have the ability to supply enough current all by itself, then you could have longevity issues with your battery.

[ogden]

I might have a partially (or hardly) charged battery while the other is fully charge.

Note that 1A current per cell, diodes will dissipate 0.5W per cell. When cells are nearly empty (at 2.5V per cell), 0.5V drop reduces efficiency of your battery pack to 75%!! All this due to your inability/laziness to fully charge cells before assembly.

[Siwastaja]

The parallel connection of course, by definition, means that each cell will be at exact same terminal voltage, at all times. Due to different series resistances within cells, they do not share current equally during high-current peaks, but then, after the peak declines, equalize each other, eventually reaching the same SoC, assuming they have the same SoC vs. voltage curve.

Note that if they are of different brand/type, and/or aged very differently, their voltage curves may differ quite a lot at some point of state-of-charge curve. This means that current sharing may differ more than you expect from just internal resistance differences alone. This is meaningless if you derate the current properly. For slow discharge and charge rates (say, around 0.2C?), there likely isn't any problem.

Of course, for the worst case, you could just derate the max current to match what the weakest cell can do alone, but this is extremely conservative.

And, laptop cells tend to have quite similar characteristics.

Just remember to match the voltages carefully (within 50-100mV) when you initially do connect them in parallel, and make that connection properly fixed so they can't accidentally drift apart, then connect again.

If you see someone claiming that li-ion cells can't or shouldn't be connected in parallel, or that diodes are "needed", ignore everything they say. They have never seen a battery pack.

[Rick Law]

I might have a partially (or hardly) charged battery while the other is fully charge.

Note that 1A current per cell, diodes will dissipate 0.5W per cell. When cells are nearly empty (at 2.5V per cell), 0.5V drop reduces efficiency of your battery pack to 75%!! All this due to your inability/laziness to fully charge cells before assembly.

That my setup uses randomly selected cells from the "charge battery" box is one of the design requirements I started with.  I keep a box of charged 18650's each installed with battery protection module - typically for my flashlights.  They are normally charged with a commercially available 18650 charger.  So, when I need it, my "power bank" always is fully charged (with cells picked from the box) instead of itself needing to be juiced up first.

Had the operational designed been permanently soldered-together fixed cells, I too would agree to use carefully match cells, solder them together with fuse protection instead of my using battery holders and cells with "battery protection modules".

I actually begun with a two matched cell with fuse protected connected in series instead of parallel, and buck down rather than boost up to 5V out.  Seeing multiple articles (& youtube videos) how some super-bright flashlights (with >2 in series) blew, I never felt comfortable with it and ended up abandoning that one.

[macboy]

Parallel is fine, if they are all the same chemistry (same nominal full/empty voltage).

I would definitely bring them to the same voltage prior to connecting them together. Unlike some other chemistries, which don't take a significant charge current until the terminal voltage is made to be quite a bit higher than the standing voltage, a Lithium cell will sink a significant current from any source higher in voltage than itself.

Parallel is much safer for mixed capacity than series. With any series connected battery, you must worry about the balance of all cells in order to prevent any one from discharging too low or charging too high, either of which is dangerous. When in parallel, all cells are at the same voltage, and therefore at (nominally) the same state of charge at all times - it is self-balancing by nature.

edit: I'll add to stay away from large charge or discharge currents. The differing internal resistances of the cells can contribute to differing cell voltage, when a large current is involved. If the current is kept relatively small (i.e. so that the V=IR drop of the internal resistances is small) then cell balance is naturally maintained.

[Rick Law]

The cold irony is, the larger the capacity the more danger they are.  We are carrying larger and larger bombs in our pocket just so we don't have to recharge our phones twice a day.

I would love to have a cell like Terminator 2 and 3 portrait - run the robot for 120 years on two cells.  But I would hate to see how it blows up, at least as portrait in T3.  I actually doubt Hydrogen Fuel Cell (as portrait in T3) could deliver that much energy. That they are fuel cells implying they get energy from the molecular bonds.  One probably have to go down to atomic level to get that much energy to keep Arnold running around for over 100 years.