Questions and advice on building a li-ion parallel 18650 battery pack

[Subterranean13]

Hi, I'm very interested in building a battery pack of my own due to the difficulty and cost of getting hold of prebuilt packs where I am (UK) and also due to personal interest and involvement. However I have a few questions left to make sure I am doing this as safely and as practically as possible that I could not find through doing a search of these forums.

Firstly, I will only be making packs of 18650 cells in parallel; I will never have any cells or sets of cells in series. My idea is to use good quality LG 2600mah unprotected cells with solder tabs, with a protection circuit such as this one, all shrink wrapped. I will likely use 3 or 4 cells to start with, but I may think about larger packs later, again always in parallel.

I have heard from people making series battery packs that one problem is making sure each cell stays balanced so as not to allow other cells to overcharge or have other problems. Are there any similar issues when making a parallel pack like I am wanting to? So long as I make sure each cell is matched in voltage when I connect them together, is there any problems that may happen over time? If one of the cells starts to die, is the result likely to be catastrophe or simply a drop in runtime?

Any help or useful links you can give are appreciated. Also if anyone has done this themselves any first hand advice would be great.

 

[liteitup]

there is someone that probably knows more then me on this subject but ill throw my 2 cents in...

Being in parallel means that you cant isolate a single cell unless you connect some type of diode before the common connection. I tore apart a dell laptop battery back to harvest some of the cells. The pack was a 4s2p 14.8v. The internal wiring of the pack had 4 "balancing" connectors going to each 2 cells connected in parallel. So in short, the 2 cells that were connected in parallel were charged together and discharged together as one. The batteries being connected in parallel will keep those batteries balanced together and act as one.

What happens when one cell goes bad i cant tell you though... it will make the parallel pack lose amp hours but will it hurt other cells connected to it by sucking down the good cell even more? iim not sure..

 

[Subterranean13]

I suspected that each 'set' of cells in parallel (in my case just the one 'set') could be treated as an individual cell, from bits pieced together from this board and also the fact that all protection circuits I've seen have been designed for a number of cells in series, never parallel. For example I've seen 1s, 2s, 3s etc but never 2p, 3p etc.

I guess you could say the max discharge current rating of a protection circuit indicates how many cells in parallel you should use it with, for example a circuit rated for 8A max discharge would never be used for a single cell, or 2 in parallel, given that each cell is only rated at 2A each.

So is it basically correct to say that cells in parallel will balance each other, and allow you to always charge them together, as one, safely? (using a proper charger of course).

 

[SilverFox]

Hello Subterranean13,

Welcome to CPF.

Not only is it basically correct, it is correct...

Tom

 

[Subterranean13]

Thanks very much guys, your help has put my mind at rest quite a bit. I will definitely be making myself one of these now. The only thing I wish I could do differently would be to be able to connect the cells directly with ultrasonic welding, but I guess that will be impossible for a single individual hobbyist. Soldering the tabs it is then.

 

[win734440]

Anyone looking for a good HID flashlight , with a goog price tag check out the HID store, they will meet or beat anyone's price. I bought a 35 watt HID and I never seen one cheaper and the quality and service rates with the best, they have a new 35 watt HID rated at 3400 lumens My next toy

 

[VidPro]

there is someone that probably knows more then me on this subject but ill throw my 2 cents in...

Being in parallel means that you cant isolate a single cell unless you connect some type of diode before the common connection.

So in short, the 2 cells that were connected in parallel were charged together and discharged together as one. The batteries being connected in parallel will keep those batteries balanced together and act as one.

What happens when one cell goes bad i cant tell you though... it will make the parallel pack lose amp hours but will it hurt other cells connected to it by sucking down the good cell even more? iim not sure..

Yes when any single one cell goes bad in a parellel pack the whole pack does not work well. if one cell self discharges , it will discharge the parellel cell connected to it. it will also put out some heat from the one cell that wont charge, with the ONE cell that wont charge taking on the whole amp load of the charge. so if a person used crappy batteries and one goes bad in the parelell set then they proceed to charge at a high rate, the rate for the parellel set going into ONE cell, could cause the one cell to heat way more than normal, even to the point of failure. due to the high charge rate going now to only one cell item.
that is why it is good to parellel with Good batteries insted of some junk from a deal site
with good cells dying Normally over long times the parellel units act much more similar, so by the time any one cell is bad, the whole pack is basically useless anyways. with cells that are inconsistant and pre-maturely fail, you could avoid high charge rates that would push one over.

 

[Therrin]

This is just the thread I was looking for!
I've been scanning through threads for a couple hours, not wanting to post up a new thread if it's been covered already.

As I've mentioned in other posts, I have about sixty 18650H's that I got from a friend(unprotected). And I realize the problem with using "cheap" cells (these were free:nana, but his company had used them for test modules, and many didn't even get used at all. And c'mon, it's hard to pass up wanting to do stuff with sixty free 18650's right?

So... Here are a few questions.

1. How do I pick relatively well "matched" cells? Check to see that they'll charge to max, and then what?

2. When charging, does the ~0.7C rate applies to the pack TOTAL? If I have 4 2200mAh cells in parallel, 8800mAh total, can I actually charge it at 6A? (I'm currently charging 2P packs @ 1.8A)

3. The protection circuits I had ordered are the SAME ones the OP listed. They list 2A amp draw cuttoff, and 2.5V as the low volt cutoff. With many cells in parallel, do you still just use ONE of these circuits for the whole pack? Or do you want to wire in one for each cell in the pack?
Also, can you drawn a higher amperage on a pack in parallel than you would on a single cell?

4. I just got a Turnigy Accucel 6 charger to charge my packs in the field. I noticed in the "Li-ion primer" this wasn't a listed suggested brand, but I see lots of people using them. Good? Bad? Indifferent?

5. Since I use my packs to power headlamps used for technical mine exploration, I am often underground for 2 to 10hrs at a time (and occasionally more with multi-day trips where I sleep inside the mine). In an effort to make my packs relatively well protected and waterproof, I've started using these holders, http://www.batteryspace.com/Battery-holder-Li-Ion-18650-Battery-Holder-1S4P-With-2.6-long-20AWG.aspx , then I wrap them in super88, put a small plastic shield across the exposed cell face, then wrap the whole thing with super88, then shrink wrap, then do two layers of plasti-dip, then shrink wrap.
Here's one that I finished last night:

- So the question is... Do I have to worry at all while charging a fully sealed pack?

If anyone can answer any of these questions, I'd really appreciate the input Thanks!

 

[SilverFox]

Hello Therrin,

To match cells you need to discharge them and view the discharge graphs. You want cells that are similar in capacity and that are similar in holding voltage during load.

Charging battery packs is done based upon the capacity of the pack. Your math looks good.

In a computer there is one protection circuit for the whole battery pack. As long as all the cells age the same, this works well. The problem comes when cells age differently. Now if you are using all the cells in parallel, they will self balance and you should be good to go.

The 0.7 charge rate is low enough that there should be no problems charging a sealed pack.

Tom

 

[AnAppleSnail]

(these were free), but his company had used them for test modules, and many didn't even get used at all. And c'mon, it's hard to pass up wanting to do stuff with sixty free 18650's right?

Nothing at all. In fact, the more 18650s the better. I think the limiting factor on bigger packs is being able to charge them...

2. When charging, does the ~0.7C rate applies to the pack TOTAL? If I have 4 2200mAh cells in parallel, 8800mAh total, can I actually charge it at 6A? (I'm currently charging 2P packs @ 1.8A)

Yes. In fact, at TOO LOW of a charge rate, charge termination can miss, or you can actually overcharge a pack. Normally the fast charge stage crams voltage up near 4.2v, then a finishing charge finishes the job, but some cheap chargers go wonky down near 0.1C charge rate. You want balanced cells or else your weak cells will interfere with fully charging the strong ones.

3. The protection circuits I had ordered are the SAME ones the OP listed. They list 2A amp draw cuttoff, and 2.5V as the low volt cutoff. With many cells in parallel, do you still just use ONE of these circuits for the whole pack? Or do you want to wire in one for each cell in the pack?
Also, can you drawn a higher amperage on a pack in parallel than you would on a single cell?

You can draw higher current from a bigger pack in parallel, as long as the cells are balanced well. If not, then drawing current will kill some cells and then you're pulling the same current from the few good cells. Using a 2A cutoff circuit will limit you to 2A, and if you're okay with that then one is fine. I do not know if the 2A limit applies to charging, but 1.8A<2A.

Hi Therrin. How's the runtime/voltage on that pack?

 

[MikeAusC]

. . . . In fact, at TOO LOW of a charge rate, charge termination can miss, or you can actually overcharge a pack. Normally the fast charge stage crams voltage up near 4.2v, then a finishing charge finishes the job, but some cheap chargers go wonky down near 0.1C charge rate. . . .

For NiMH yes, but not for LiIon.

For LiIon approaching full charge the Charger is in Constant Voltage Mode and the current flowing is determined by the battery.

A good LiIon charger will sense the current dropping to a low level and disconnect all voltage from the cell to maximise cell life - but a basic CC-CV charger will work as well to preserve battery life, if you remove the cells after 24 hours. I've been doing that for years.

 

[Therrin]

Silverfox,
Hey thanks for your response.

You mentioned watching the "discharge graphs"... how is that done? You mean on a computer?

One of the options in my Turnigy 6 charger says something about usb, but I don't see a usb plug on it. The instructions mention something about a computer program for it but it didn't come with anything.

What kind of charger/discharger do you use?

AnAppleSnail,
APPLES!! Hey man! Good to see you around. I was doing some solo work last night, down in a mine and noticed my HP20 that this pack is connected to is pretty dim. And it should be giving full output. The HP20 runs on 6v and I've been using 1S2P 18650 with it for a while now. The other night while I was letting a new pack discharge it, it felt like it was getting really hot just sitting still, so I set it outside where it's cold.
After I brought it in I thought it was dim just cuz I removed the regulated output circuitry from the light when I removed the battery pack, but now even with a fully charged pack it stays dim. Maybe 80lm or so of output instead of 220.

Any idea what could have caused that? Could the LED have overheated and be running at reduced capacity? It's also a lot "bluer" light now than it used to be.

I'm tempted to pull the back out of it and see about replacing the XPG with an XML

 

[AnAppleSnail]

The other night while I was letting a new pack discharge it, it felt like it was getting really hot just sitting still, so I set it outside where it's cold.
After I brought it in I thought it was dim just cuz I removed the regulated output circuitry from the light when I removed the battery pack, but now even with a fully charged pack it stays dim. Maybe 80lm or so of output instead of 220.

Any idea what could have caused that? Could the LED have overheated and be running at reduced capacity? It's also a lot "bluer" light now than it used to be.

I'm tempted to pull the back out of it and see about replacing the XPG with an XML

Sounds like you cooked it. The XM-L will be floodier, of course.

 

[SilverFox]

Hello Therrin,

I use a CBA from West Mountain Radio to monitor discharge performance. I use a Schulze charger for charging.

Tom

 

[Therrin]

Sounds like you cooked it. The XM-L will be floodier, of course.

Damn... I took the back plate off earlier today. There's a .. hmm... I should take a picture. It doesn't look very easy to mod. I wish I could figure out how to get the hinge feet out of the plastic frame without breaking it.

 

[Therrin]

Gonna start building some 1S4P packs now.

Subterranean13, hows your stuff coming along?

 

[Changchung]

Hi, let me ask you guys, can I use differents mah batteries connect it them in parallel?


SFMI4UT

 

[SilverFox]

Hello Changchung

With Li-Ion chemistry, yes.

Tom

 

[Changchung]

Thanks Tom...


SFMI4UT