Small pack (20 Cell) help

Thatbakerguy

New member
Joined
Jan 9, 2019
Messages
12
Hope this is the right forum. I've read through the FAQ forum as much as I could and after I got out of there I have these questions.

Project overview:

I have a module that requires 12-24v and 20 18650 batteries that are going to go in a tote able case. After reading the FAQ it appears that 18650's or Li-Ion in generalare not ideal for a 12v system. at least in a 3s configuration.

Would I be better off conforming to a 24v system or a 4s 12v system? The hardware can handle the 12-16.8v range. I am concerned that the 24v system would be too close to the upper edge to be safe.

I think at most the system would pull 5A a peak.

as far as battery pack design goes it seems that the general setup would be charger>bms> 18650batteries in xS configuration.


Some other general questions are:




[*]Does a BMS protect against depleting the batteries as in if I leave the moduleon does it have protection against draining the batteries too low. (I've had previous issues with NiCad)
[*]BMS recommendations. my project is pretty simple i believe the batrium system would be overkill.
[*]Charger recommendations TP4056/TP5100?
 
If your hardware can do up to 16.8V then sticking with a "12V" 4S5P config would avoid converters & associated extra losses.
4x 4.1V (max full state charge for long life) would be 16.4V
You also say the module does up to 24V??
Is the 20x 18650 cells a size/cost limit?
If you did a 24V pack it's 7S so 21 cells in 7S3P would be the number.

If you pick a simple 4S BMS it should be fine for "12V" - look for more balancing current if you can find it.
Take care to use repacker to get the pack's capacity as even between the each of the "P groups of 5 cells.
Also look for a BMS with a low voltage cut-off feature or maybe add a separate LVCO module.

5A peak would be 1A/cell assuming this is at 12V.
 
Thanks for the reply.


The 20 Cell limit is just what i have on hand.


I think I agree with you i would like to stay away from converters to avoid the power loss there.

as far as BMS is concerned is an ebay module sufficient? like this onedoesntseem to have LVCO.

How did you arrive at the 1A/cell number at a 5A peak? What is an acceptable amperage to draw at a peak? Is the 1A high or low in this regard?

Thanks again!
 
BMS: Totally depends on usage. If unsupervised 7x24 operation with used cells is a thing then you probably need a BMS. Occasional use for a limited time while you are present and you probably can live without a BMS. Same if you use new cells. Also depends on how you charge the cells and if the devices you will run from the battery have their own LVCO.

Peak Amps: 5A peak on a 4s5p battery -> 1A peak per cell. 1A is very low for most lithium cells in good condition. Acceptable peak amps are whatever you have tested (in case of used cells) or what the specs say (in case of new cells). There is no general rule and the spread between different cells is huge. Some cells can't really deal with high peak loads, other cells will do 100A peak without too much struggle. It also depends on your definition of peak. Peak is generally everything that isn't continuous, but a peak could be 0.5s or 5s and that makes a huge difference.
 
DarkRaven said:
BMS: Totally depends on usage. If unsupervised 7x24 operation with used cells is a thing then you probably need a BMS. Occasional use for a limited time while you are present and you probably can live without a BMS. Same if you use new cells. Also depends on how you charge the cells and if the devices you will run from the battery have their own LVCO.

Peak Amps: 5A peak on a 4s5p battery -> 1A peak per cell. 1A is very low for most lithium cells in good condition. Acceptable peak amps are whatever you have tested (in case of used cells) or what the specs say (in case of new cells). There is no general rule and the spread between different cells is huge. Some cells can't really deal with high peak loads, other cells will do 100A peak without too much struggle. It also depends on your definition of peak. Peak is generally everything that isn't continuous, but a peak could be 0.5s or 5s and that makes a huge difference.

So the cells that i have are new "Ultra Fire" ebay 18650's. i understand their reliability may be suspect.my goal is to have this setup be put into operation at any time and un-supervised so Im trying to make it as complete as possible i dont want to have to baby sit it. The devices I will be power from them dont have an LVCO.
 
no "may be" here, they "are". You will never have a successful bms setup with "Fire" anything branded cells. They will always be out of balance as no 2 cells are the same.
 
Korishan said:
no "may be" here, they "are". You will never have a successful bms setup with "Fire" anything branded cells. They will always be out of balance as no 2 cells are the same.

What brands should I be looking for. I don't need top of the line for my project but what is reliable.
 
I was thinking great thread until the part regarding the cells

Send them back or Throw them away

Spend under $40 to get some decent cells that have been reclaimed

Ultrafire is junk

Go with a name brand reclaimed.....
 
Either go for reclaimed cells, if you have a source, and test them or buy new cells, for example Samsung if they should be 18650 or 21700. But you might also decide to use other sizes where other manufacturers come into play, it depends on what you want to achieve in terms of capacity.
 
Have to agree, Ultrafire anything > bin.
Also avoid Sanyo red jacketed cells, they're usually "heaters".

You mentioned your peak draw was 5A so this is divided by the number of cells in parallel.
Your pack (assuming 4S5P) has 4 sets of 5 cells in parallel (see the FAQ page on how it works).
So 5A / 5 cells in parallel = 1A/cell
 
A bms is a battery management system and can contain some or all of your requirements

You would need
1. Over voltage protection
2. Under voltage protection
3. balancing
4. Cell level protection

This is minium
with this above and a PSU that have a decent current and CV limit at around 16.8 or such will work just fine.
Example of BMS that would work
 
Not just "Ultra" fire. ANY of the "Fire" branded cells steer clear of. Ultra is just the most common. It has been found for some of them to be filled with sand and there really be a small AA lithium cell inside. So going with any of the "Fire" branded cells you are gambling, and with high stakes.
 
Chablis_m said:
I was thinking great thread until the part regarding the cells

Send them back or Throw them away

Spend under $40 to get some decent cells that have been reclaimed

Ultrafire is junk

Go with a name brand reclaimed.....

Hey I can admit i went the cheap route.

i'll see if i can find some to re-claim or a dependable source of name brand.

In the mean time I think i've found a suitable BMS. but I'm still stumped on a charge. What should i be looking for. I'm assuming I'll need something around 14.8v or do i need to go up to 16.4?
 
You need to get 16.4 or so volts or you won't have much capacity available!
 
Thatbakerguy said:
Chablis_m said:
.......but I'm still stumped on a charge. What should i be looking for. I'm assuming I'll need something around 14.8v or do i need to go up to 16.4?

Buy a variable power supply, you'll not regret it.

Zero to thirty volts or more,a couple of amps or more, with constant current and voltage control.
 
I just got my hands on some old 18v power tool batteries (makita). I'm going to try and pull the batteries out of them.

Can someone link me a testing procedure?

I didnt see anything in the faq.

Thanks,
 
You test them just like any other cell. Opus, liito, imax, etc testers.
 
quick update and questions

all cells measured over 4.2v after sitting for a week or so.

I picked up a imax b6 to use for the discharge testing. according to the spec sheet my cells specify a minimum voltage of 2.5v but the b6 only goes down to 3v.

I've discharged two cells and they are both showing ~1300ma which would make sense if they are 1500ma rated.

Does it hurt to only dicharge to 3v vs 2.5?
 
Back
Top