Powerbank battery configuration

[IgorM]

I want to make a powerbank using laptop battery cells (18650) and my question is how should I configure it? I'm not sure if it should be 12v using cells in series and how many of them can be connected in parallel? Is a 3s4p a safe option? 

[Electronics Wizardy]

What do you want in terms of output voltage and current?

 

What controller are you using?

[Tabs]

12 minutes ago, IgorM said:

I want to make a powerbank using laptop battery cells (18650) and my question is how should I configure it? I'm not sure if it should be 12v using cells in series and how many of them can be connected in parallel? Is a 3s4p a safe option? 

Whenever you have a series/parallel configuration you run the risk of one cell deteriorating much faster than others (age, heat, humidity, or simply wear or manufacturing inconsistencies) which means that when charging the full pack you run the risk of reverse charging a cell in a series, which is super dangerous. This is especially prevalent if you mix and match cells to create the pack.

 

The safest configuration would be to have every cell monitored and charged directly and individually from a controller, to ensure that no one cell is in a direct circuit with another cell. This can however get expensive if you need a lot of current like @Electronics Wizardy suggested, since it means you'll need a lot of cells controlled.

 

Until we know what load you need, it's hard to say how well a 3s4p config would work. With all cells perfectly balanced that would give you between 11.2 and 12.6v at up to 3A (the highest I've seen from a single generic 18650 is 1.3A so I'm being conservative).

[IgorM]

12 minutes ago, Electronics Wizardy said:

What do you want in terms of output voltage and current?

 

What controller are you using?

i'm planning to get 12 cells (2 laptop Batteries)  but not sure how to arrange them

[IgorM]

4 minutes ago, Tabs said:

Whenever you have a series/parallel configuration you run the risk of one cell deteriorating much faster than others (age, heat, humidity, or simply wear or manufacturing inconsistencies) which means that when charging the full pack you run the risk of reverse charging a cell in a series, which is super dangerous. This is especially prevalent if you mix and match cells to create the pack.

 

The safest configuration would be to have every cell monitored and charged directly and individually from a controller, to ensure that no one cell is in a direct circuit with another cell. This can however get expensive if you need a lot of current like @Electronics Wizardy suggested, since it means you'll need a lot of cells controlled.

 

Until we know what load you need, it's hard to say how well a 3s4p config would work. With all cells perfectly balanced that would give you between 11.2 and 12.6v at up to 3A (the highest I've seen from a single generic 18650 is 1.3A so I'm being conservative).

not going to use it for higher than 3 or even 2 amps. For a controller I'm planning to get a 3s bms with balance feature if I create a 3s4p pack. My question is if I can better arrange about 12 cells and if using 4 in parallel is fine.

[Tabs]

10 minutes ago, IgorM said:

not going to use it for higher than 3 or even 2 amps. For a controller I'm planning to get a 3s bms with balance feature if I create a 3s4p pack. My question is if I can better arrange about 12 cells and if using 4 in parallel is fine.

For that configuration, using a 3s4p is totally fine mate, just make sure each individual cell is able to be charged to the same voltage before you assemble the pack (and that all cells are charged the same). Do you have a decent multimeter? The voltages don't have to be exact between cells, but +-0.1v is perfect in a series, and the average between parallel arrays needs to be between +-0.05v.

[IgorM]

3 minutes ago, Tabs said:

For that configuration, using a 3s4p is totally fine mate, just make sure each individual cell is able to be charged to the same voltage before you assemble the pack (and that all cells are charged the same). Do you have a decent multimeter? The voltages don't have to be exact between cells, but +-0.1v is perfect in a series, and the average between parallel arrays needs to be between +-0.05v.

I have a basic multimeter that should be able to check that. But considering that the cells are from a laptop battery they are in a 3s2p config and should be around the same voltage.

[Tabs]

3 minutes ago, IgorM said:

I have a basic multimeter that should be able to check that. But considering that the cells are from a laptop battery they are in a 3s2p config and should be around the same voltage.

Please check them though mate, here's a link to a video by a youtuber called Big Clive where he dissected a laptop battery that failed because one cell in 6 was bad - it ended up being reverse charged even though it was part of the same pack. Better to be safe than sorry if you work with high current lithium cells.

 

link - 

 

[Hackentosher]

Go on ebay and get a diy power bank case for $4

[Unimportant]

4 hours ago, IgorM said:

not going to use it for higher than 3 or even 2 amps. For a controller I'm planning to get a 3s bms with balance feature if I create a 3s4p pack. My question is if I can better arrange about 12 cells and if using 4 in parallel is fine.

Do note that most commercial controllers/balancers are not going to work very good when mixing cells.

Normally, when a battery pack is assembled in the factory, all the cells are pretty much equal, coming from the same batch with equal characteristics. That means they'll charge and discharge about evenly by themselves. The balancer circuits are only designed to handle small imbalances that arise over time, with many cycles.

 

You'll find that if you start mixing and matching totally different cells from 2 battery packs the balancer is going to have a tough time keeping that balanced and is probably going to give up pretty quickly. Even if the cells from the 2 packs are the same make and type, chances are they're not the same age and have a different wear level.

 

I designed my own, ultra aggressive, balancer circuit to handle this. Basically charging all the cells in series with a fixed current and having a bypass transistor in parallel with each cell that bypasses the amount of current required to keep the cell from ever going above 4.2V. When all cells are bypassed the pack is charged.

 

Is this a 5V USB powerbank you're trying to make ? If so, the best option would be to simply put all your cells in parallel, to make 1 big 3.7V battery. Charging/discharging cells in parallel is no problem (the parallel pack behaves the same then a single cell), it's only series usage that's problematic. You can then use a boost converter to boost the 3.7V to 5V.