Building a 18650 battery pack - advice needed

9 posts / 0 new
Last post
Sirius9
Sirius9's picture
Offline
Last seen: 7 hours 18 min ago
Joined: 03/04/2012 - 19:23
Posts: 4288
Location: South Europe
Building a 18650 battery pack - advice needed

I have some cells that I pulled from laptop battery packs, they are just sitting in the box waiting their

destiny, so I thought I should put them to use. I decided to make couple battery packs,

mostly cells in parallel setup to get higher capacity, I have PCB for power bank that I could use for this purpose.

Here is a list showing individual cell capacity measured while discharge cicle at 500mA:

 bts

Cells marked with blue as well as cell marked in green I would put in parallel and from cells marked in yellow

I would make 8.4V pack (2P3S). This one I would use for my bike lights.

I have no idea what to do with cells with small capacity (first 12 cells in the table: 1011mAh to 1593mAh).

Any thoughts guys!

 

Thanks

https://docs.google.com/spreadsheets/d/1oyeKjYqGcPBST4RVZdTwcmZUqWjTyvVR...

 

bilakos10
bilakos10's picture
Online
Last seen: 7 min 22 sec ago
Joined: 08/22/2013 - 14:25
Posts: 419
Location: Thessaloniki, Greece

The best thing you could do with these cells is to use them in low wattage applications.
You can create some DIY garden lights. Connect a led strip to a couple of paralleled cells and hook them up to a cheap solar panel.

My daily deals thread: ☢ [Gearbest.com Special Deals]
If you need any coupon code, just leave a reply on the thread.

Sirius9
Sirius9's picture
Offline
Last seen: 7 hours 18 min ago
Joined: 03/04/2012 - 19:23
Posts: 4288
Location: South Europe

Yes that would be interesting (althou cheap solar panels are mostly crap) but unfortunately I live in building apartment so no garden :/

 

Ronin42
Ronin42's picture
Offline
Last seen: 1 day 7 hours ago
Joined: 12/17/2014 - 22:17
Posts: 1627
Location: Alameda, CA

One trick I learned is to bottom balance the cells.

Its less important when all the cells are topped up than it is to know that all the cells run out at the same time. (for series arrangements)

heck use the first 12 P (approx 15-16 ah) and make your own uber power bank to learn and abuse (does not matter what happens to them).

To charge just run (2 wires) 1 each with a wire with two magnets from you regular charger to the battery bank (one +, one-) and let it treat it like a giant 18650 cell.

(“It’s good that most people can’t remember their previous lives. Otherwise
things would be a lot more complicated than they already are.”
Ajaan Lee Dhammadharo)

dekozn
Offline
Last seen: 22 hours 2 min ago
Joined: 06/17/2016 - 09:04
Posts: 579
Location: Be

How do you bottom balance cells? I’ve heard about it but nevver really got to know how to do it right.

JakeDjanitor
Offline
Last seen: 9 hours 43 min ago
Joined: 10/27/2015 - 09:00
Posts: 725
Location: North East USA

Ronin42 wrote:
One trick I learned is to bottom balance the cells.

Its less important when all the cells are topped up than it is to know that all the cells run out at the same time. (for series arrangements)

heck use the first 12 P (approx 15-16 ah) and make your own uber power bank to learn and abuse (does not matter what happens to them).

To charge just run (2 wires) 1 each with a wire with two magnets from you regular charger to the battery bank (one +, one-) and let it treat it like a giant 18650 cell.

What happens if one cell get a to 4.2 volts and others are at 3.9? Does it stop charging?
Or 3 are at 4.2 and one at 3.9 will the other 3 over heat and be dangerous?

gauss163
Offline
Last seen: 6 hours 34 min ago
Joined: 10/09/2013 - 18:04
Posts: 750
Location: USA
Ronin42 wrote:
One trick I learned is to bottom balance the cells […]

Bottom-balancing is generally not a good idea. For an introductory exposition on cell balancing techniques see this exposition by TI’s guru.

FreeMagenta
Offline
Last seen: 1 hour 23 min ago
Joined: 10/12/2014 - 12:50
Posts: 158
Location: Poland

JakeDjanitor wrote:
Ronin42 wrote:
One trick I learned is to bottom balance the cells.

Its less important when all the cells are topped up than it is to know that all the cells run out at the same time. (for series arrangements)

heck use the first 12 P (approx 15-16 ah) and make your own uber power bank to learn and abuse (does not matter what happens to them).

To charge just run (2 wires) 1 each with a wire with two magnets from you regular charger to the battery bank (one +, one-) and let it treat it like a giant 18650 cell.

What happens if one cell get a to 4.2 volts and others are at 3.9? Does it stop charging?
Or 3 are at 4.2 and one at 3.9 will the other 3 over heat and be dangerous?

If the cells are in parrarell and there are two cells with same capacity but significant internal resistance difference (IR for short -say, 50 mOhm and 450 mOhm), while unloaded, voltages are equal. In the moment a current is forced into the cell, in first seconds most of the current flows through the lower resistance one. However, since it means that 50 mOhm cell is being charged faster, its voltage increase quicker , while on 500 mOhm it is going up more slowly.

Let’s say both batteries are at 3,2V and then 1A current charging starts. at the beginning, the resistance is 9 times greater on 450 mOhm, so there is only 0,1A, while 50 mOhm gets 0,9A. So the batteries resist actively with 3,2V + passive IR.
After a while 50 mOhm one got more charge, so its no-load voltage increased to 3,25V, while 450 one got less charge and have only 3,22V. Thus, the charging current flows 0,84A through 50 mOhm and 0,16A through 450 mOhm cell. When the voltage difference reaches 0,4V, the current equalizes on both batteries.
Now, when charging goes into CV stage, with 4,2V the 50 mOhm cell is topping up quickly (it has over 4,17V of no-load voltage(, while 450 mOhm still has 3,8V, so current on 50 mOhm quickly decreases while current on 450 mOhm is fading slowly. Since the cutoff should happen at charge current/10 the 50 will be a little overcharged and 450 a little undercharged. After cutoff, the voltages on batteries are not equal, so the 50 will charge 450 a little, until voltage equalizes.

The bad thing is – that at the start ‘better’ cell gets practically full current charge, so if we have to charge cell with 0.5A, for a while better cell is charged with too high curent.

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 56 min ago
Joined: 02/19/2014 - 14:46
Posts: 1970
Location: Alhama de Murcia, Spain

Balancing is not really an issue if you use a BMS with balancing and/or a few balance boards, affordable stuff in AliExpress.

If you can live with a little bit more losses you can use independent 1S BMSes per stage. By using a multi-pole switch you can route each stage anode(s) to each B/P+ BMS input, in such a way that when the switch is off (all BMSes powered down) all of the MOSFETs disconnect and the series P-/anode links are disengaged. Additionally, if we bridge all of the switch pins for the “off” position (all anodes connected) we would just need to bridge the cathodes for a self-balancing full parallel setup, doable, for example, with a multi-pole 5V relay (or more). Now add some TP4056 boards and you can even charge via multiple in parallel phone/tablet chargers. 

No errors are to be found in this dissertation.

 

Cheers Smile